{ "cells": [ { "cell_type": "markdown", "id": "8073f4d2", "metadata": {}, "source": [ "# Study 003 · Can conservative alignment improve calibration?\n", "\n", "Daybreak Neural · 8 November 2026\n", "\n", "**Completed exploratory development study.** At a budget of 24 original calibration trials, blended alignment plus updating reached **67.47%** mean balanced accuracy, versus **67.47%** for updating alone. The paired difference was **+0.0067 percentage points**, with a conditional 95% run-bootstrap interval of **−0.67 to +0.61 points**. This study did not demonstrate an added benefit from alignment on the primary comparison.\n", "\n", "The conservative rule mitigated the previously observed eight-trial failure for participant 3, while worsening results for two other participants relative to full alignment. The hypothesis was motivated by Study 002 outcomes on these same people and sessions. **Participants 5–9 remain reserved.**\n", "\n", "All 180 scores were independently recomputed from 40,185 prediction rows (893 distinct test trials), all 25 interval rows were reconciled, and all 31,255 predictions for the seven unchanged methods exactly reproduced Study 002." ] }, { "cell_type": "markdown", "id": "ff465365", "metadata": {}, "source": [ "## Context and methods\n", "\n", "Study 002 found a short-prefix failure of full covariance alignment. Here we test one fixed conservative rule: `lambda = n / (n + 48)`, where `n` is the accepted calibration count, and `R_blend = (1 - lambda) R_source + lambda R_prefix`. Target covariance matrices are whitened using the symmetric inverse square root of this blend. The source tangent reference and scaler stay fixed.\n", "\n", "The constant 48 is an engineering choice, not an optimized or calibrated prior. With 24 accepted trials the target estimate receives one-third of the covariance weight. No alternative strengths are tried in this experiment.\n", "\n", "**Primary question:** does blended alignment plus the same supervised update outperform update-only at 24 original trials? Secondary comparisons include the full-alignment combination at eight and 24 trials. We retain every original Study 002 method unchanged and require exact replication of its saved predictions.\n", "\n", "**Data access:** all accepted day-one task trials fit each person's models. The first day-two task run supplies the original chronological prefixes 0, 8, 16, 24 and 48. Artifact exclusions follow those boundaries. The same five later runs provide accepted test trials at every budget. No later recording enters a fit or shared preprocessing reference. However, analysts have already seen these sessions' results in prior studies: this is development evidence, not fresh validation.\n", "\n", "Each prediction uses a four-second imagery epoch. Source artifact flags define an offline accepted-trial analysis. Neither a positive point estimate nor a conditional interval establishes live performance, calibration-time savings or scientific novelty." ] }, { "cell_type": "markdown", "id": "f50f696d", "metadata": {}, "source": [ "**Execution record.** The 27 original code cells were executed top-to-bottom in a fresh Python namespace, with tables and three figures saved in this notebook. This environment blocks kernel sockets, so an in-process runner was used; the interactive Jupyter interface was not tested here. The final reference-check cell was then executed against the saved outputs. Independent scripts reconciled all scores, selected bootstrap intervals and unchanged Study 002 predictions. No model was refitted to write this interpretation." ] }, { "cell_type": "markdown", "id": "4db1c496", "metadata": {}, "source": [ "### Setup and portable rerun\n", "\n", "Select **Python 3 (ipykernel)**. The definitions are included below, so this notebook can run independently of the accompanying engine file. Install the bundle's pinned `requirements.txt` in a separate Python environment for the closest reproduction. Missing dependencies can also be installed with `%pip install numpy scipy pandas scikit-learn mne pyriemann matplotlib threadpoolctl`.\n", "\n", "The first run may download eight public MATLAB files. Cached source files are checked against recorded SHA-256 hashes. Only participants 1–4 can be loaded. `NEURAL_DATA_DIR` optionally points to an existing cache; otherwise `eeg_data` is used. Results go to `results` beside the notebook's working directory. No account or API key is required." ] }, { "cell_type": "code", "execution_count": 1, "id": "3c379fae", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Data: /workspace/scratch/b15a265229b0/tmp/eeg_data\n", "Results: /workspace/scratch/b15a265229b0/conservative_alignment_study/results\n" ] } ], "source": [ "from pathlib import Path\n", "import os, json, hashlib\n", "from IPython.display import display\n", "from threadpoolctl import threadpool_limits\n", "DATA_DIRECTORY = Path(os.environ.get('NEURAL_DATA_DIR', 'eeg_data')).expanduser().resolve()\n", "OUTPUT_DIRECTORY = Path('results').resolve()\n", "OUTPUT_DIRECTORY.mkdir(parents=True, exist_ok=True)\n", "print('Data:', DATA_DIRECTORY)\n", "print('Results:', OUTPUT_DIRECTORY)" ] }, { "cell_type": "markdown", "id": "09c81874", "metadata": {}, "source": [ "### Fixed protocol and implementation\n", "\n", "The timestamp records when this design was fixed locally, before Study 003 outputs. It does not make the development study an external preregistration. Supervised updating uses the unchanged Study 002 weights: target fraction `n/(n+48)` and total fitting weight equal to the source sample count. This classifier weighting is separate from the covariance blend, even though both use the same numerical count rule.\n", "\n", "Expand the definition cells to inspect the loader, fitting paths, blending and paired whole-run bootstrap. Source preprocessing and all seven reference methods remain unchanged." ] }, { "cell_type": "code", "execution_count": 2, "id": "63e46d34", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Design fixed: 2026-11-08T20:12:59.558558+00:00\n", "Primary: {'budget': 24, 'contrast': 'blended_combined minus update_only', 'endpoint': 'equal-participant mean balanced accuracy', 'unit': 'percentage points'}\n", "Blend: n_accepted/(n_accepted+48)\n", "Development participants: [1, 2, 3, 4]\n" ] } ], "source": [ "FROZEN_PROTOCOL = {'study': '003', 'frozen_at_utc': '2026-11-08T20:12:59.558558+00:00', 'scope': 'failure-motivated exploratory development on previously inspected participants and sessions; not an independent validation or external preregistration', 'subjects': [1, 2, 3, 4], 'reserved_subjects': [5, 6, 7, 8, 9], 'budgets_original_trials': [0, 8, 16, 24, 48], 'source_training': 'all accepted day-one task trials', 'calibration': 'accepted subset of original chronological prefix of first day-two task run', 'test': 'all accepted trials in day-two task runs2–6, fixed for all budgets/methods', 'primary': {'budget': 24, 'contrast': 'blended_combined minus update_only', 'endpoint': 'equal-participant mean balanced accuracy', 'unit': 'percentage points'}, 'bootstrap_replicates': 5000, 'bootstrap_seed': 20261109, 'target_mixture_weight': 'n_accepted / (n_accepted + 48)', 'total_sample_weight': 'N_source', 'filter': 'Butterworth4thorder8–30Hz causal sosfilt peroriginaltrial [0,6)s, crop [2,6)s; no inter-trial state', 'lr': {'C': 1.0, 'solver': 'lbfgs', 'max_iter': 3000, 'tol': 1e-06, 'class_weight': None}, 'csp': {'n_components': 8, 'reg': 'oas', 'cov_est': 'epoch', 'rank': 'full', 'log': True, 'norm_trace': False, 'component_order': 'mutual_info'}, 'covariance': 'OAS pertrial, followed by covariance congruence for alignment variant', 'tangent_space': {'metric': 'riemann', 'tsupdate': False, 'reference': 'source-only'}, 'methods': ['bandpower_LDA', 'CSP_LDA', 'raw_tangent_LR', 'frozen', 'alignment_only', 'update_only', 'combined', 'blended_alignment', 'blended_combined'], 'adaptation_setting': 'fixed at end of prefix; no later online updates', 'source_hashes': {'A01E.mat': '53d415f39c3d7b0c88b894d7b08d99bcdfe855ede63831d3691af1a45607fb62', 'A01T.mat': '054f02e70cf9c4ada1517e9b9864f45407939c1062c6793516585c6f511d0325', 'A02E.mat': 'd63c454005d3a9b41d8440629482e855afc823339bdd0b5721842a7ee9cc7b12', 'A02T.mat': '5ddd5cb520b1692c3ba1363f48d98f58f0e46f3699ee50d749947950fc39db27', 'A03E.mat': 'd4229267ec7624fa8bd3af5cbebac17f415f7c722de6cb676748f8cb3b717d97', 'A03T.mat': '7e731ee8b681d5da6ecb11ae1d4e64b1653c7f15aad5d6b7620b25ce53141e80', 'A04E.mat': '81916dff2c12997974ba50ffc311da006ea66e525010d010765f0047e771c86a', 'A04T.mat': '15850d81b95fc88cc8b9589eb9b713d49fa071e28adaf32d675b3eaa30591d6e'}, 'prior_study_engine_sha256': '531f5ccf190f589701be652c9d10a0d61f3a50c693b2fa213028d70f905d4e7e', 'alignment_blend': {'lambda': 'n_accepted/(n_accepted+48)', 'matrix': '(1-lambda)*mean_source_covariance + lambda*mean_accepted_prefix_covariance', 'space': 'original EEG covariance coordinates before whitening', 'transform': 'symmetric inverse square root; congruence transform of existing OAS covariance', 'strength': 48, 'interpretation': 'single fixed heuristic; no search or claim of calibrated prior strength'}, 'secondary': [{'budget': 8, 'contrast': 'blended_combined minus combined'}, {'budget': 24, 'contrast': 'blended_combined minus combined'}, {'budget': 24, 'contrast': 'blended_combined minus frozen'}, {'budget': 24, 'contrast': 'blended_combined minus blended_alignment'}], 'descriptive_diagnostics': ['per-participant changes relative to frozen at each budget', 'known participant3 eight-trial failure comparison', 'target blend strength and covariance diagnostic checks', 'leave-one-participant-out primary mean'], 'no_tuning': 'No parameter or budget selection using Study003 results; do not evaluate participants5–9.'}\n", "print('Design fixed:', FROZEN_PROTOCOL['frozen_at_utc'])\n", "print('Primary:', FROZEN_PROTOCOL['primary'])\n", "print('Blend:', FROZEN_PROTOCOL['alignment_blend']['lambda'])\n", "print('Development participants:', FROZEN_PROTOCOL['subjects'])" ] }, { "cell_type": "code", "execution_count": 3, "id": "2298ffc7", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "\"\"\"Study 003 conservative covariance alignment benchmark; participants 1--4 only.\n", "\n", "Existing-method ablation, not a novel-algorithm claim. Prediction occurs after\n", "the complete four-second epoch. The fixed n/(n+48) blend was chosen after\n", "Study002 development results and before Study003 results. No parameter search\n", "or held-out participant evaluation is performed.\n", "\"\"\"\n", "\n", "from __future__ import annotations\n", "\n", "import argparse\n", "\n", "import hashlib\n", "\n", "import json\n", "\n", "import platform\n", "\n", "import time\n", "\n", "import urllib.request\n", "\n", "import warnings\n", "\n", "from datetime import datetime, timezone\n", "\n", "from pathlib import Path\n", "\n", "import mne\n", "\n", "import numpy as np\n", "\n", "import pandas as pd\n", "\n", "import pyriemann\n", "\n", "import scipy\n", "\n", "import sklearn\n", "\n", "from mne.decoding import CSP\n", "\n", "from pyriemann.estimation import Covariances\n", "\n", "from pyriemann.tangentspace import TangentSpace\n", "\n", "from scipy.integrate import trapezoid\n", "\n", "from scipy.io import loadmat\n", "\n", "from scipy.signal import butter, sosfilt, welch\n", "\n", "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n", "\n", "from sklearn.exceptions import ConvergenceWarning\n", "\n", "from sklearn.linear_model import LogisticRegression\n", "\n", "from sklearn.metrics import accuracy_score, balanced_accuracy_score\n", "\n", "from sklearn.preprocessing import StandardScaler\n", "\n", "SUBJECTS = (1, 2, 3, 4)\n", "\n", "BUDGETS = (0, 8, 16, 24, 48)\n", "\n", "LABELS = (1, 2, 3, 4)\n", "\n", "CLASS_NAMES = (\"left hand\", \"right hand\", \"feet\", \"tongue\")\n", "\n", "METHODS = (\"bandpower_LDA\", \"CSP_LDA\", \"raw_tangent_LR\", \"frozen\",\n", " \"alignment_only\", \"update_only\", \"combined\",\n", " \"blended_alignment\", \"blended_combined\")\n", "\n", "ABLATION_METHODS = (\"frozen\", \"alignment_only\", \"update_only\", \"combined\",\n", " \"blended_alignment\", \"blended_combined\")\n", "\n", "BLEND_PRIOR_TRIALS = 48.0\n", "\n", "SOURCE_URL = \"https://bnci-horizon-2020.eu/database/data-sets/001-2014/\"\n", "\n", "EXPECTED_SHA256 = {\n", " \"A01E.mat\": \"53d415f39c3d7b0c88b894d7b08d99bcdfe855ede63831d3691af1a45607fb62\",\n", " \"A01T.mat\": \"054f02e70cf9c4ada1517e9b9864f45407939c1062c6793516585c6f511d0325\",\n", " \"A02E.mat\": \"d63c454005d3a9b41d8440629482e855afc823339bdd0b5721842a7ee9cc7b12\",\n", " \"A02T.mat\": \"5ddd5cb520b1692c3ba1363f48d98f58f0e46f3699ee50d749947950fc39db27\",\n", " \"A03E.mat\": \"d4229267ec7624fa8bd3af5cbebac17f415f7c722de6cb676748f8cb3b717d97\",\n", " \"A03T.mat\": \"7e731ee8b681d5da6ecb11ae1d4e64b1653c7f15aad5d6b7620b25ce53141e80\",\n", " \"A04E.mat\": \"81916dff2c12997974ba50ffc311da006ea66e525010d010765f0047e771c86a\",\n", " \"A04T.mat\": \"15850d81b95fc88cc8b9589eb9b713d49fa071e28adaf32d675b3eaa30591d6e\",\n", "}" ] }, { "cell_type": "code", "execution_count": 4, "id": "6d932dff", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _require(condition, message):\n", " if not bool(condition):\n", " raise AssertionError(message)" ] }, { "cell_type": "code", "execution_count": 5, "id": "045e15fd", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _sha256(path):\n", " digest = hashlib.sha256()\n", " with Path(path).open(\"rb\") as stream:\n", " for chunk in iter(lambda: stream.read(1024 * 1024), b\"\"):\n", " digest.update(chunk)\n", " return digest.hexdigest()" ] }, { "cell_type": "code", "execution_count": 6, "id": "fc212075", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _verified_source_file(data_dir, filename):\n", " \"\"\"Use the cache or atomically download one of eight allowlisted files.\n", "\n", " Downloads are capped at 128 MiB and approximately four minutes including a\n", " final blocking read (60-second per-operation timeout, 180-second deadline).\n", " A file is never promoted from .part unless its recorded hash matches.\n", " \"\"\"\n", " if filename not in EXPECTED_SHA256:\n", " raise ValueError(\"Source file is outside the development-participant allowlist\")\n", " directory = Path(data_dir)\n", " directory.mkdir(parents=True, exist_ok=True)\n", " path = directory / filename\n", " if not path.exists():\n", " temporary = directory / (filename + \".part\")\n", " started, size = time.monotonic(), 0\n", " try:\n", " with urllib.request.urlopen(SOURCE_URL + filename, timeout=60) as response:\n", " with temporary.open(\"wb\") as stream:\n", " while True:\n", " if time.monotonic() - started > 180:\n", " raise TimeoutError(f\"Download deadline exceeded: {filename}\")\n", " chunk = response.read(1024 * 1024)\n", " if not chunk:\n", " break\n", " size += len(chunk)\n", " if size > 128 * 1024 * 1024:\n", " raise ValueError(f\"Download size limit exceeded: {filename}\")\n", " stream.write(chunk)\n", " _require(_sha256(temporary) == EXPECTED_SHA256[filename],\n", " f\"Downloaded source hash mismatch: {filename}\")\n", " temporary.replace(path)\n", " finally:\n", " if temporary.exists():\n", " temporary.unlink()\n", " _require(_sha256(path) == EXPECTED_SHA256[filename], f\"Source hash mismatch: {filename}\")\n", " return path" ] }, { "cell_type": "code", "execution_count": 7, "id": "e27d0e56", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _legacy_bandpower(epoch):\n", " frequencies, psd = welch(epoch, fs=250, window=\"hann\", nperseg=250,\n", " noverlap=125, detrend=\"constant\", scaling=\"density\", axis=0)\n", " features = []\n", " for low, high in ((8.0, 12.0), (13.0, 30.0)):\n", " selected = (frequencies >= low) & (frequencies <= high)\n", " power = trapezoid(psd[selected], frequencies[selected], axis=0)\n", " _require(np.all(power > 0), \"Nonpositive bandpower\")\n", " features.extend(np.log10(power))\n", " return np.asarray(features, dtype=np.float64)" ] }, { "cell_type": "code", "execution_count": 8, "id": "7d682464", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def load_participant(subject, data_dir):\n", " \"\"\"Verified files, all-original-trial ledger, and row-aligned feature arrays.\n", "\n", " Causal filtering has zero initial state separately for every original\n", " [0,6) s trial. Crop to [2,6) s, then independently demean each channel.\n", " \"\"\"\n", " if isinstance(subject, bool) or subject not in SUBJECTS:\n", " raise ValueError(\"Only development participants 1, 2, 3, 4 are allowed\")\n", " subject = int(subject)\n", " sos = butter(4, [8.0, 30.0], btype=\"bandpass\", fs=250, output=\"sos\")\n", " rows, epochs, bandpower, files = [], [], [], []\n", " for session, suffix in ((1, \"T\"), (2, \"E\")):\n", " filename = f\"A{subject:02d}{suffix}.mat\"\n", " path = _verified_source_file(data_dir, filename)\n", " digest = _sha256(path)\n", " _require(digest == EXPECTED_SHA256[filename], f\"Source hash mismatch: {filename}\")\n", " files.append({\"filename\": filename, \"sha256\": digest,\n", " \"source_url\": SOURCE_URL + filename, \"bytes\": path.stat().st_size})\n", " runs = np.atleast_1d(loadmat(path, simplify_cells=True)[\"data\"])\n", " task_run, session_trial = 0, 0\n", " for source_run, run in enumerate(runs, 1):\n", " starts = np.asarray(run[\"trial\"], dtype=int).reshape(-1)\n", " if len(starts) == 0:\n", " continue\n", " task_run += 1\n", " labels = np.asarray(run[\"y\"], dtype=int).reshape(-1)\n", " flags = np.asarray(run[\"artifacts\"], dtype=int).reshape(-1)\n", " signal = np.asarray(run[\"X\"], dtype=np.float64)\n", " _require(int(run[\"fs\"]) == 250 and signal.shape[1] == 25, \"Unexpected EEG layout\")\n", " _require(len(starts) == len(labels) == len(flags) == 48, \"Expected 48 trials/run\")\n", " _require(tuple(run[\"classes\"]) == CLASS_NAMES, \"Unexpected class order\")\n", " _require(set(flags).issubset({0, 1}), \"Unexpected artifact flag\")\n", " for within_run, (start, label, artifact) in enumerate(zip(starts, labels, flags), 1):\n", " session_trial += 1\n", " start0 = int(start) - 1\n", " raw6 = signal[start0:start0 + 1500, :22]\n", " _require(raw6.shape == (1500, 22) and np.isfinite(raw6).all(), \"Invalid EEG epoch\")\n", " filtered = sosfilt(sos, raw6, axis=0)[500:1500].T.copy()\n", " filtered -= filtered.mean(axis=1, keepdims=True)\n", " epochs.append(filtered)\n", " bandpower.append(_legacy_bandpower(raw6[500:1500]))\n", " region = \"source\" if session == 1 else (\"calibration_pool\" if task_run == 1 else \"common_test\")\n", " rows.append({\"subject\": subject, \"session\": session, \"filename\": filename,\n", " \"source_run\": source_run, \"task_run\": task_run, \"trial_in_run\": within_run,\n", " \"trial_in_session\": session_trial,\n", " \"trial_id\": f\"s{subject:02d}_d{session}_r{task_run:02d}_t{within_run:02d}\",\n", " \"source_start_sample_1based\": int(start), \"source_start_sample_0based\": start0,\n", " \"label\": int(label), \"artifact\": int(artifact),\n", " \"accepted\": int(artifact == 0), \"region\": region})\n", " _require(task_run == 6 and session_trial == 288, \"Expected six runs and 288 trials/session\")\n", " ledger = pd.DataFrame(rows)\n", " _require(len(ledger) == 576 and ledger.trial_id.is_unique, \"Invalid ledger\")\n", " for session in (1, 2):\n", " counts = ledger.loc[ledger.session == session].groupby(\"label\").size().to_dict()\n", " _require(counts == {label: 72 for label in LABELS}, \"Unexpected original class counts\")\n", " return {\"ledger\": ledger, \"epochs\": np.asarray(epochs),\n", " \"bandpower\": np.asarray(bandpower), \"source_files\": files}" ] }, { "cell_type": "code", "execution_count": 9, "id": "1985a43e", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _invsqrt(matrix):\n", " values, vectors = np.linalg.eigh((matrix + matrix.T) / 2.0)\n", " _require(np.isfinite(values).all() and np.min(values) > 0, \"Covariance is not SPD\")\n", " floor = np.max(values) * 1e-12\n", " return (vectors * np.maximum(values, floor) ** -0.5) @ vectors.T" ] }, { "cell_type": "code", "execution_count": 10, "id": "fe829737", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _blended_reference(source_mean, target_mean, weight):\n", " \"\"\"Convex covariance blend and checked symmetric inverse square root.\n", "\n", " This is a fixed count-based engineering rule, not an estimated optimum.\n", " Endpoint checks concern numerical implementation, not decoder performance.\n", " \"\"\"\n", " _require(np.isfinite(weight) and 0.0 <= weight <= 1.0,\n", " \"Blend weight must be finite and within [0,1]\")\n", " source_whitener = _invsqrt(source_mean)\n", " target_whitener = _invsqrt(target_mean)\n", " blended = (1.0 - weight) * source_mean + weight * target_mean\n", " _require(np.isfinite(blended).all(), \"Nonfinite blended covariance\")\n", " whitener = _invsqrt(blended)\n", " _require(np.isfinite(whitener).all(), \"Nonfinite blended whitener\")\n", " endpoint_zero = (1.0 - 0.0) * source_mean + 0.0 * target_mean\n", " endpoint_one = (1.0 - 1.0) * source_mean + 1.0 * target_mean\n", " _require(np.allclose(_invsqrt(endpoint_zero), source_whitener,\n", " rtol=1e-12, atol=1e-12), \"Lambda-zero endpoint failed\")\n", " _require(np.allclose(_invsqrt(endpoint_one), target_whitener,\n", " rtol=1e-12, atol=1e-12), \"Lambda-one endpoint failed\")\n", " raw_shift = float(np.linalg.norm(target_mean - source_mean, ord=\"fro\"))\n", " blended_shift = float(np.linalg.norm(blended - source_mean, ord=\"fro\"))\n", " _require(np.isclose(blended_shift, weight * raw_shift,\n", " rtol=1e-9, atol=1e-10), \"Covariance blend contraction failed\")\n", " return blended, whitener" ] }, { "cell_type": "code", "execution_count": 11, "id": "c53c46a3", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _congruence(covariances, whitening):\n", " result = whitening @ covariances @ whitening.T\n", " return (result + result.transpose(0, 2, 1)) / 2.0" ] }, { "cell_type": "code", "execution_count": 12, "id": "d490d374", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _lr():\n", " return LogisticRegression(C=1.0, solver=\"lbfgs\", max_iter=3000,\n", " tol=1e-6, class_weight=None)" ] }, { "cell_type": "code", "execution_count": 13, "id": "b252aaf4", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _fit(estimator, X, y, audit, subject, stage, trial_ids,\n", " budget=None, sample_weight=None):\n", " \"\"\"Fail on convergence warnings; record every fitting/reference input ID.\"\"\"\n", " with warnings.catch_warnings(record=True) as captured:\n", " warnings.simplefilter(\"error\", ConvergenceWarning)\n", " with mne.use_log_level(\"ERROR\"):\n", " if sample_weight is None:\n", " estimator.fit(X, y)\n", " else:\n", " estimator.fit(X, y, sample_weight=sample_weight)\n", " entry = {\"subject\": subject, \"stage\": stage, \"budget\": budget,\n", " \"n_rows\": len(trial_ids), \"trial_ids\": list(trial_ids),\n", " \"n_iter\": np.asarray(getattr(estimator, \"n_iter_\", [])).tolist(),\n", " \"warnings\": [str(item.message) for item in captured]}\n", " if sample_weight is not None:\n", " entry[\"sample_weight_sum\"] = float(np.sum(sample_weight))\n", " audit.append(entry)\n", " return estimator" ] }, { "cell_type": "code", "execution_count": 14, "id": "076124f6", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _features(covariances, tangent, scaler):\n", " return scaler.transform(tangent.transform(covariances))" ] }, { "cell_type": "code", "execution_count": 15, "id": "87016f1f", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def _write_json(path, value):\n", " Path(path).write_text(json.dumps(value, indent=2, allow_nan=False) + \"\\n\", encoding=\"utf-8\")" ] }, { "cell_type": "code", "execution_count": 16, "id": "6e892a5e", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def run_study(data_dir, output_dir):\n", " \"\"\"Execute fixed development methods and save auditable, EEG-free results.\n", "\n", " Returns DataFrames under scores, predictions, trial_ledger,\n", " calibration_counts, uncertainty, leave_one_participant_out, blend_diagnostics; metadata is a\n", " dictionary. Each table is saved under its key plus .csv; metadata is saved\n", " as access_audit.json. Calibration counts contain one row/method/budget/person.\n", " \"\"\"\n", " output_dir = Path(output_dir)\n", " output_dir.mkdir(parents=True, exist_ok=True)\n", " audit, ledgers, calibration_counts, scores, frames, files = [], [], [], [], [], []\n", " blend_diagnostics = []\n", " for subject in SUBJECTS:\n", " print(f\"Participant {subject}/4: loading verified recordings and fitting fixed methods\", flush=True)\n", " data = load_participant(subject, data_dir)\n", " ledger, epochs, bandpower = data[\"ledger\"], data[\"epochs\"], data[\"bandpower\"]\n", " ledgers.append(ledger)\n", " files.extend(data[\"source_files\"])\n", " source = ledger.index[(ledger.region == \"source\") & (ledger.accepted == 1)].to_numpy()\n", " test = ledger.index[(ledger.region == \"common_test\") & (ledger.accepted == 1)].to_numpy()\n", " pool = ledger.index[ledger.region == \"calibration_pool\"].to_numpy()\n", " source_ids = ledger.loc[source, \"trial_id\"].tolist()\n", " test_ids = ledger.loc[test, \"trial_id\"].tolist()\n", " ys, yt = ledger.loc[source, \"label\"].to_numpy(), ledger.loc[test, \"label\"].to_numpy()\n", " _require(set(ys) == set(LABELS) and set(yt) == set(LABELS), \"Missing source/test class\")\n", " _require(set(ledger.loc[test, \"task_run\"]) == {2, 3, 4, 5, 6}, \"Expected later five test runs\")\n", " _require(len(pool) == 48 and not set(source_ids).intersection(test_ids), \"Source/test split failure\")\n", "\n", " def fit(estimator, X, y, stage, ids=source_ids, budget=None, sample_weight=None):\n", " return _fit(estimator, X, y, audit, subject, stage, ids, budget, sample_weight)\n", "\n", " bp_scaler = fit(StandardScaler(), bandpower[source], None, \"bandpower_scaler\")\n", " bp_model = fit(LinearDiscriminantAnalysis(solver=\"lsqr\", shrinkage=\"auto\"),\n", " bp_scaler.transform(bandpower[source]), ys, \"bandpower_LDA\")\n", " fixed = {\"bandpower_LDA\": bp_model.predict(bp_scaler.transform(bandpower[test]))}\n", " csp = fit(CSP(n_components=8, reg=\"oas\", cov_est=\"epoch\", rank=\"full\", log=True,\n", " norm_trace=False, component_order=\"mutual_info\"), epochs[source], ys, \"CSP\")\n", " csp_model = fit(LinearDiscriminantAnalysis(solver=\"lsqr\", shrinkage=\"auto\"),\n", " csp.transform(epochs[source]), ys, \"CSP_LDA\")\n", " fixed[\"CSP_LDA\"] = csp_model.predict(csp.transform(epochs[test]))\n", "\n", " # Each covariance depends only on that trial; no cross-test fitting.\n", " covariance = Covariances(estimator=\"oas\")\n", " cs = covariance.transform(epochs[source])\n", " ct = covariance.transform(epochs[test])\n", " raw_tangent = fit(TangentSpace(metric=\"riemann\", tsupdate=False), cs, None, \"raw_tangent_reference\")\n", " raw_source = raw_tangent.transform(cs)\n", " raw_scaler = fit(StandardScaler(), raw_source, None, \"raw_tangent_scaler\")\n", " raw_lr = fit(_lr(), raw_scaler.transform(raw_source), ys, \"raw_tangent_LR\")\n", " fixed[\"raw_tangent_LR\"] = raw_lr.predict(_features(ct, raw_tangent, raw_scaler))\n", "\n", " source_mean = cs.mean(axis=0)\n", " ws = _invsqrt(source_mean)\n", " audit.append({\"subject\": subject, \"stage\": \"source_whitening_reference\", \"budget\": None,\n", " \"n_rows\": len(source), \"trial_ids\": source_ids, \"labels_used\": False})\n", " aligned_cs = _congruence(cs, ws)\n", " _require(np.allclose(aligned_cs.mean(axis=0), np.eye(22), atol=1e-7), \"Source recentering failed\")\n", " tangent = fit(TangentSpace(metric=\"riemann\", tsupdate=False), aligned_cs, None, \"ablation_tangent_reference\")\n", " source_features = tangent.transform(aligned_cs)\n", " scaler = fit(StandardScaler(), source_features, None, \"ablation_scaler\")\n", " zs = scaler.transform(source_features)\n", " base_lr = fit(_lr(), zs, ys, \"ablation_source_LR\")\n", " unaligned_test = _features(_congruence(ct, ws), tangent, scaler)\n", " frozen_prediction = base_lr.predict(unaligned_test)\n", "\n", " for budget in BUDGETS:\n", " prefix = pool[:budget]\n", " calibration = prefix[ledger.loc[prefix, \"accepted\"].to_numpy(dtype=bool)]\n", " ncal = len(calibration)\n", " cal_ids = ledger.loc[calibration, \"trial_id\"].tolist()\n", " yc = ledger.loc[calibration, \"label\"].to_numpy()\n", " _require(not set(cal_ids).intersection(test_ids), \"Calibration/test overlap\")\n", " _require(all(ledger.loc[calibration, \"trial_in_run\"] <= budget), \"Prefix budget exceeded\")\n", " class_counts = {f\"class_{label}_accepted\": int(np.sum(yc == label)) for label in LABELS}\n", " elapsed = 0.0 if budget == 0 else float(\n", " (ledger.loc[prefix[-1], \"source_start_sample_0based\"] - ledger.loc[prefix[0], \"source_start_sample_0based\"]) / 250 + 6.0)\n", " pred = dict(fixed)\n", " pred[\"frozen\"] = frozen_prediction\n", " iterations = {\"bandpower_LDA\": None, \"CSP_LDA\": None,\n", " \"raw_tangent_LR\": int(np.max(raw_lr.n_iter_)), \"frozen\": int(np.max(base_lr.n_iter_))}\n", " alpha = ncal / (ncal + 48.0) if ncal else 0.0\n", " blend_weight = ncal / (ncal + BLEND_PRIOR_TRIALS) if ncal else 0.0\n", " if ncal == 0:\n", " for method in (\"alignment_only\", \"update_only\", \"combined\",\n", " \"blended_alignment\", \"blended_combined\"):\n", " pred[method] = frozen_prediction.copy()\n", " iterations[method] = int(np.max(base_lr.n_iter_))\n", " blend_diagnostics.append({\"subject\": subject, \"budget\": budget,\n", " \"accepted_trials\": 0, \"blend_lambda\": 0.0,\n", " \"target_relative_frobenius_shift\": None,\n", " \"blended_relative_frobenius_shift\": 0.0,\n", " \"source_condition_number\": float(np.linalg.cond(source_mean)),\n", " \"target_condition_number\": None,\n", " \"blended_condition_number\": float(np.linalg.cond(source_mean))})\n", " else:\n", " cc = covariance.transform(epochs[calibration])\n", " target_mean = cc.mean(axis=0)\n", " wc = _invsqrt(target_mean)\n", " audit.append({\"subject\": subject, \"stage\": \"target_alignment_reference\", \"budget\": budget,\n", " \"n_rows\": ncal, \"trial_ids\": cal_ids, \"labels_used\": False})\n", " aligned_test = _features(_congruence(ct, wc), tangent, scaler)\n", " pred[\"alignment_only\"] = base_lr.predict(aligned_test)\n", " iterations[\"alignment_only\"] = int(np.max(base_lr.n_iter_))\n", " blended_mean, wb = _blended_reference(source_mean, target_mean, blend_weight)\n", " audit.append({\"subject\": subject, \"stage\": \"blended_alignment_reference\",\n", " \"budget\": budget, \"n_rows\": len(source) + ncal,\n", " \"trial_ids\": source_ids + cal_ids, \"labels_used\": False,\n", " \"accepted_target_trials\": ncal, \"blend_lambda\": blend_weight})\n", " blended_test = _features(_congruence(ct, wb), tangent, scaler)\n", " _require(np.isfinite(blended_test).all(), \"Nonfinite blended test features\")\n", " pred[\"blended_alignment\"] = base_lr.predict(blended_test)\n", " iterations[\"blended_alignment\"] = int(np.max(base_lr.n_iter_))\n", " denominator = float(np.linalg.norm(source_mean, ord=\"fro\"))\n", " blend_diagnostics.append({\"subject\": subject, \"budget\": budget,\n", " \"accepted_trials\": ncal, \"blend_lambda\": blend_weight,\n", " \"target_relative_frobenius_shift\": float(np.linalg.norm(target_mean - source_mean, ord=\"fro\") / denominator),\n", " \"blended_relative_frobenius_shift\": float(np.linalg.norm(blended_mean - source_mean, ord=\"fro\") / denominator),\n", " \"source_condition_number\": float(np.linalg.cond(source_mean)),\n", " \"target_condition_number\": float(np.linalg.cond(target_mean)),\n", " \"blended_condition_number\": float(np.linalg.cond(blended_mean))})\n", " weights = np.r_[np.full(len(source), 1.0 - alpha), np.full(ncal, alpha * len(source) / ncal)]\n", " _require(np.isclose(weights.sum(), len(source)), \"Supervised weights must sum to Ns\")\n", " for method, whitening, test_features in (\n", " (\"update_only\", ws, unaligned_test), (\"combined\", wc, aligned_test),\n", " (\"blended_combined\", wb, blended_test)):\n", " zc = _features(_congruence(cc, whitening), tangent, scaler)\n", " updated = fit(_lr(), np.vstack([zs, zc]), np.r_[ys, yc], method,\n", " ids=source_ids + cal_ids, budget=budget, sample_weight=weights)\n", " pred[method] = updated.predict(test_features)\n", " iterations[method] = int(np.max(updated.n_iter_))\n", " if budget == 0:\n", " _require(all(np.array_equal(pred[\"frozen\"], pred[m]) for m in ABLATION_METHODS),\n", " \"Zero-budget ablation predictions differ\")\n", " for method in METHODS:\n", " labels_used = ncal if method in (\"update_only\", \"combined\", \"blended_combined\") else 0\n", " target_used = ncal if method in (\"alignment_only\", \"update_only\", \"combined\",\n", " \"blended_alignment\", \"blended_combined\") else 0\n", " method_blend = blend_weight if method in (\"blended_alignment\", \"blended_combined\") else 0.0\n", " alignment_weight = (1.0 if ncal else 0.0) if method in (\"alignment_only\", \"combined\") else method_blend\n", " calibration_counts.append({\"subject\": subject, \"method\": method, \"budget\": budget,\n", " \"available_original_trials\": len(prefix), \"accepted_trials\": ncal,\n", " \"flagged_exclusions\": len(prefix) - ncal,\n", " \"target_calibration_trials_used\": target_used, \"calibration_labels_used\": labels_used,\n", " \"blend_lambda\": method_blend, \"alignment_target_weight\": alignment_weight,\n", " \"all_four_classes_present\": set(yc) == set(LABELS),\n", " \"prefix_elapsed_seconds_to_last_epoch_end\": elapsed, **class_counts})\n", " scores.append({\"subject\": subject, \"method\": method, \"budget\": budget,\n", " \"n_source\": len(source), \"n_test\": len(test), \"accepted_calibration_available\": ncal,\n", " \"target_calibration_trials_used\": target_used, \"calibration_labels_used\": labels_used,\n", " \"blend_lambda\": method_blend, \"alignment_target_weight\": alignment_weight,\n", " \"supervised_target_total_weight_fraction\": alpha if labels_used else 0.0,\n", " \"accuracy\": float(accuracy_score(yt, pred[method])),\n", " \"balanced_accuracy\": float(balanced_accuracy_score(yt, pred[method])),\n", " \"correct\": int(np.sum(yt == pred[method])), \"classifier_n_iter\": iterations[method]})\n", " frame = ledger.loc[test, [\"subject\", \"session\", \"task_run\", \"trial_in_run\",\n", " \"trial_in_session\", \"trial_id\", \"label\"]].copy()\n", " frame[\"prediction\"], frame[\"method\"], frame[\"budget\"] = pred[method], method, budget\n", " frames.append(frame)\n", "\n", " allowed = set(source_ids) | set(ledger.loc[pool, \"trial_id\"])\n", " for entry in (record for record in audit if record[\"subject\"] == subject):\n", " _require(set(entry[\"trial_ids\"]).issubset(allowed), \"Fit accessed unallowed data\")\n", " _require(not set(entry[\"trial_ids\"]).intersection(test_ids), \"Common test trial in fit audit\")\n", " if entry.get(\"budget\") is not None:\n", " allowed_prefix = set(ledger.loc[pool[:entry[\"budget\"]], \"trial_id\"])\n", " _require(set(entry[\"trial_ids\"]).issubset(set(source_ids) | allowed_prefix),\n", " \"Fit access exceeded original prefix budget\")\n", "\n", " score_frame = pd.DataFrame(scores)\n", " predictions = pd.concat(frames, ignore_index=True)\n", " uncertainty, leave_one_out = summarize_uncertainty(predictions, score_frame)\n", " tables = {\"scores\": score_frame, \"predictions\": predictions,\n", " \"trial_ledger\": pd.concat(ledgers, ignore_index=True),\n", " \"calibration_counts\": pd.DataFrame(calibration_counts),\n", " \"uncertainty\": uncertainty, \"leave_one_participant_out\": leave_one_out,\n", " \"blend_diagnostics\": pd.DataFrame(blend_diagnostics)}\n", " for name, table in tables.items():\n", " table.to_csv(output_dir / f\"{name}.csv\", index=False)\n", " metadata = {\n", " \"study\": \"003\", \"dataset\": \"BNCI2014_001\", \"created_utc\": datetime.now(timezone.utc).isoformat(),\n", " \"subjects_read\": list(SUBJECTS), \"reserved_subjects_never_read\": [5, 6, 7, 8, 9],\n", " \"budgets_original_chronological_trials\": list(BUDGETS), \"methods\": list(METHODS),\n", " \"primary_comparison\": \"blended_combined minus update_only at 24 original prefix trials\",\n", " \"secondary_comparisons\": [\"blended_combined minus combined at 8\", \"blended_combined minus combined at 24\",\n", " \"blended_combined minus frozen at 24\", \"blended_combined minus blended_alignment at 24\"],\n", " \"blend_rule\": {\"target_weight\": \"n/(n+48), n accepted calibration trials\", \"prior_trial_constant\": BLEND_PRIOR_TRIALS,\n", " \"covariance\": \"(1-lambda)*source_mean + lambda*prefix_mean\",\n", " \"whitener\": \"symmetric inverse square root of blended covariance\",\n", " \"selection\": \"No within-Study003 tuning or parameter search; Study002 development outcomes motivated this rule\"},\n", " \"source_files\": files,\n", " \"versions\": {\"python\": platform.python_version(), \"numpy\": np.__version__, \"pandas\": pd.__version__,\n", " \"scipy\": scipy.__version__, \"scikit_learn\": sklearn.__version__,\n", " \"mne\": mne.__version__, \"pyriemann\": pyriemann.__version__},\n", " \"access_rules\": {\n", " \"source\": \"All accepted day-one trials, independently per participant\",\n", " \"calibration\": \"Accepted trials within original first-day-two-run prefix; no replacement of flagged trials\",\n", " \"test\": \"Same accepted trials in day-two task runs 2--6 for all methods and budgets\",\n", " \"test_covariances\": \"Each OAS covariance constructed independently from the current trial; no cross-test estimation\",\n", " \"tangent_and_scaler\": \"Source-fitted once per representation; tsupdate=False; unchanged at every budget\",\n", " \"target_alignment\": \"Mean OAS covariance of accepted prefix only; OAS not recomputed after congruence\",\n", " \"preprocessing\": \"250 Hz, first 22 EEG; causal fourth-order Butterworth 8--30 Hz over [0,6), zero state per trial; crop [2,6), demean per channel\",\n", " \"bandpower\": \"Study001 unfiltered raw [2,6) Welch 44 log10 bandpower features\",\n", " \"supervised_weights\": \"alpha=n/(n+48); each source 1-alpha; each target alpha*Ns/n; sum Ns\",\n", " \"blended_alignment\": \"Convex source/available-prefix OAS covariance mean with lambda=n/(n+48); source tangent and scaler unchanged\",\n", " \"zero_budget\": \"All six ablation arms use source whitening and the same frozen classifier\"},\n", " \"bootstrap\": {\"repetitions\": 5000, \"seed\": 20261109, \"unit\": \"whole test task run\",\n", " \"conditional_on\": \"These four participants, their two sessions, observed calibration prefixes, and fitted models\",\n", " \"aggregate\": \"Equal mean participant BA; independently resampled runs within each fixed participant\",\n", " \"limitations\": \"No participant resampling, refitting, new-day uncertainty, or confirmatory population inference; unadjusted percentile intervals\"},\n", " \"fit_access_audit\": audit,\n", " \"engine_sha256\": _sha256(__file__) if \"__file__\" in globals() and Path(__file__).is_file() else None,\n", " }\n", " _write_json(output_dir / \"access_audit.json\", metadata)\n", " return {**tables, \"metadata\": metadata}" ] }, { "cell_type": "code", "execution_count": 17, "id": "eeeaf37f", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def summarize_uncertainty(predictions, scores=None, repetitions=5000, seed=20261109):\n", " \"\"\"Paired conditional whole-run bootstrap for prespecified contrasts.\n", "\n", " Within-person run draws are reused across methods and budgets. Aggregate\n", " means use four fixed people, independently resampling their observed runs.\n", " Only blended-combined versus update-only at budget24 is primary.\n", " \"\"\"\n", " contrasts = {\n", " 8: ((\"combined\", \"secondary\"),),\n", " 24: ((\"update_only\", \"primary\"), (\"combined\", \"secondary\"),\n", " (\"frozen\", \"secondary\"), (\"blended_alignment\", \"secondary\")),\n", " }\n", " _require(set(predictions.subject) == set(SUBJECTS),\n", " \"Exactly four development people required\")\n", " rng = np.random.default_rng(seed)\n", " run_draws = {subject: rng.integers(0, 5, size=(repetitions, 5))\n", " for subject in SUBJECTS}\n", " points, bootstraps, rows = {}, {}, []\n", "\n", " def append_interval(subject, budget, point, bootstrap):\n", " for comparator, role in contrasts[budget]:\n", " ci, mi = METHODS.index(\"blended_combined\"), METHODS.index(comparator)\n", " low, high = np.quantile(bootstrap[:, ci] - bootstrap[:, mi], [.025, .975])\n", " rows.append({\"subject\": subject, \"budget\": budget,\n", " \"contrast\": f\"blended_combined_minus_{comparator}\",\n", " \"prespecified_role\": role,\n", " \"estimate_pp\": float(100 * (point[ci] - point[mi])),\n", " \"ci_low_pp\": float(100 * low), \"ci_high_pp\": float(100 * high),\n", " \"repetitions\": repetitions,\n", " \"interval_type\": \"conditional paired whole-run percentile; unadjusted\"})\n", "\n", " for budget in contrasts:\n", " for subject in SUBJECTS:\n", " person = predictions.loc[(predictions.subject == subject) &\n", " (predictions.budget == budget)]\n", " paired = person.pivot(index=[\"task_run\", \"trial_id\", \"label\"],\n", " columns=\"method\", values=\"prediction\").reset_index()\n", " _require(paired[list(METHODS)].notna().all().all(), \"Unpaired predictions\")\n", " run_ids = sorted(paired.task_run.unique())\n", " _require(run_ids == [2, 3, 4, 5, 6], \"Expected five common test runs\")\n", " totals = np.zeros((5, 4), dtype=int)\n", " correct = np.zeros((5, len(METHODS), 4), dtype=int)\n", " for ri, run in enumerate(run_ids):\n", " for ci, label in enumerate(LABELS):\n", " group = paired.loc[(paired.task_run == run) & (paired.label == label)]\n", " totals[ri, ci] = len(group)\n", " for mi, method in enumerate(METHODS):\n", " correct[ri, mi, ci] = int(np.sum(group[method] == label))\n", " _require((totals > 0).all(), \"A test run lacks a class\")\n", " draws = run_draws[subject]\n", " boot_ba = (correct[draws].sum(axis=1) /\n", " totals[draws].sum(axis=1)[:, None, :]).mean(axis=2)\n", " point_ba = (correct.sum(axis=0) / totals.sum(axis=0)[None, :]).mean(axis=1)\n", " points[(subject, budget)], bootstraps[(subject, budget)] = point_ba, boot_ba\n", " if scores is not None:\n", " observed = scores.loc[(scores.subject == subject) &\n", " (scores.budget == budget)].set_index(\"method\")\n", " _require(np.allclose(point_ba, observed.loc[list(METHODS), \"balanced_accuracy\"]),\n", " \"Bootstrap point and observed score differ\")\n", " append_interval(subject, budget, point_ba, boot_ba)\n", " append_interval(\"mean\", budget,\n", " np.mean([points[(subject, budget)] for subject in SUBJECTS], axis=0),\n", " np.mean([bootstraps[(subject, budget)] for subject in SUBJECTS], axis=0))\n", " ci, mi = METHODS.index(\"blended_combined\"), METHODS.index(\"update_only\")\n", " leave_one_out = [{\"omitted_subject\": omitted, \"budget\": 24,\n", " \"contrast\": \"blended_combined_minus_update_only\",\n", " \"estimate_pp\": float(100 * np.mean([points[(s, 24)][ci] - points[(s, 24)][mi]\n", " for s in SUBJECTS if s != omitted])),\n", " \"interpretation\": \"descriptive sensitivity; not independent validation\"}\n", " for omitted in SUBJECTS]\n", " return pd.DataFrame(rows), pd.DataFrame(leave_one_out)" ] }, { "cell_type": "markdown", "id": "6f5bed8c", "metadata": {}, "source": [ "## Execute the fixed comparison\n", "\n", "All methods use the same accepted later-run trials. Six adaptation variants have identical zero-budget predictions. Source hashes, covariance endpoints, finite features, access boundaries, supervised weights and convergence are checked during execution." ] }, { "cell_type": "code", "execution_count": 18, "id": "28c44360", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Participant 1/4: loading verified recordings and fitting fixed methods\n", "Participant 2/4: loading verified recordings and fitting fixed methods\n", "Participant 3/4: loading verified recordings and fitting fixed methods\n", "Participant 4/4: loading verified recordings and fitting fixed methods\n", "Study003 completed.\n" ] } ], "source": [ "with threadpool_limits(limits=1):\n", " study_result = run_study(DATA_DIRECTORY, OUTPUT_DIRECTORY)\n", "print('Study003 completed.')" ] }, { "cell_type": "code", "execution_count": 19, "id": "df970d6f", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
budget08162448
subject
108162245
208162446
307142246
407132035
\n", "
" ], "text/plain": [ "budget 0 8 16 24 48\n", "subject \n", "1 0 8 16 22 45\n", "2 0 8 16 24 46\n", "3 0 7 14 22 46\n", "4 0 7 13 20 35" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
budget08162448
subject
10.00.1430.2500.3140.484
20.00.1430.2500.3330.489
30.00.1270.2260.3140.489
40.00.1270.2130.2940.422
\n", "
" ], "text/plain": [ "budget 0 8 16 24 48\n", "subject \n", "1 0.0 0.143 0.250 0.314 0.484\n", "2 0.0 0.143 0.250 0.333 0.489\n", "3 0.0 0.127 0.226 0.314 0.489\n", "4 0.0 0.127 0.213 0.294 0.422" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "Accepted calibration counts by original budget:\n", "Actual target covariance weight:\n" ] } ], "source": [ "import pandas as pd\n", "scores = pd.read_csv(OUTPUT_DIRECTORY / 'scores.csv')\n", "predictions = pd.read_csv(OUTPUT_DIRECTORY / 'predictions.csv')\n", "calibration = pd.read_csv(OUTPUT_DIRECTORY / 'calibration_counts.csv')\n", "intervals = pd.read_csv(OUTPUT_DIRECTORY / 'uncertainty.csv')\n", "blend_diagnostics = pd.read_csv(OUTPUT_DIRECTORY / 'blend_diagnostics.csv')\n", "print('Accepted calibration counts by original budget:')\n", "display(calibration.drop_duplicates(['subject','budget']).pivot(index='subject', columns='budget', values='accepted_trials'))\n", "print('Actual target covariance weight:')\n", "display(blend_diagnostics.pivot(index='subject', columns='budget', values='blend_lambda').round(3))" ] }, { "cell_type": "markdown", "id": "1b1a2f85", "metadata": {}, "source": [ "## Results\n", "\n", "Balanced accuracy is mean recall over four classes. Points at different budgets use the same later test trials. The curves are descriptive; they are not a search for the best budget. Participant-level plots show whether an average hides different responses." ] }, { "cell_type": "code", "execution_count": 20, "id": "1167827c", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "\"\"\"Portable plotting for Study 003; inputs are reviewed result tables.\"\"\"\n", "\n", "from pathlib import Path\n", "\n", "import numpy as np\n", "\n", "import pandas as pd\n", "\n", "import matplotlib.pyplot as plt\n", "\n", "DISPLAY_LABELS = {'bandpower_LDA':'Bandpower + LDA','CSP_LDA':'CSP + LDA','raw_tangent_LR':'Raw tangent + LR', 'frozen':'Frozen tangent decoder','alignment_only':'Alignment only','update_only':'Supervised update only','combined':'Full alignment + update','blended_alignment':'Blended alignment only','blended_combined':'Blended alignment + update'}\n", "\n", "COLORS = {'frozen':'#5b6571','alignment_only':'#b58119','update_only':'#276a9f','combined':'#ae4c36','blended_combined':'#207056'}\n", "\n", "MARKERS = {'frozen':'s','alignment_only':'^','update_only':'o','combined':'D','blended_combined':'^'}\n", "\n", "ARMS=['frozen','update_only','combined','blended_combined']" ] }, { "cell_type": "code", "execution_count": 21, "id": "55ea11dc", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def configure_style():\n", " plt.rcParams.update({'font.family':'DejaVu Sans','font.size':11,'axes.titlesize':13,'axes.labelsize':11,'axes.spines.top':False,'axes.spines.right':False,'savefig.dpi':160,'figure.facecolor':'white','axes.facecolor':'white'})" ] }, { "cell_type": "code", "execution_count": 22, "id": "7f63470d", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def calibration_curves(scores, output_dir):\n", " configure_style(); output_dir=Path(output_dir); output_dir.mkdir(parents=True,exist_ok=True)\n", " fig,axes=plt.subplots(2,2,figsize=(12,8),sharex=True,sharey=True)\n", " for subject,ax in zip([1,2,3,4],axes.flat):\n", " for method in ARMS:\n", " rows=scores.query('subject == @subject and method == @method').sort_values('budget')\n", " ax.plot(rows.budget,100*rows.balanced_accuracy,label=DISPLAY_LABELS[method],color=COLORS[method],marker=MARKERS[method],linewidth=1.8,markersize=5,linestyle='--' if method=='frozen' else '-')\n", " ax.axhline(25,color='#aaaaaa',linestyle=':',linewidth=1)\n", " ax.set(title=f'Participant {subject}',ylim=(0,100),xticks=[0,8,16,24,48],yticks=[0,25,50,75,100])\n", " ax.grid(axis='y',alpha=.18)\n", " for ax in axes[1]:ax.set_xlabel('Original calibration-trial budget')\n", " for ax in axes[:,0]:ax.set_ylabel('Balanced accuracy (%)')\n", " handles,labels=axes[0,0].get_legend_handles_labels()\n", " fig.legend(handles,labels,loc='lower center',bbox_to_anchor=(.5,.045),ncol=2,frameon=False)\n", " fig.suptitle('Study 003: calibration budget and later-session performance',fontsize=16,y=.98)\n", " fig.text(.5,.015,'Development participants 1–4 · Same accepted test trials in runs 2–6 at every budget\\nLines connect tested budgets; they do not imply guaranteed improvement. Dashed horizontal reference: 25%.',ha='center',fontsize=9,color='#555555')\n", " fig.subplots_adjust(left=.09,right=.98,top=.9,bottom=.2,wspace=.13,hspace=.25)\n", " fig.savefig(output_dir/'calibration_curves.png',bbox_inches='tight')\n", " return fig" ] }, { "cell_type": "code", "execution_count": 23, "id": "87910f44", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def aggregate_curves(scores, output_dir):\n", " configure_style();output_dir=Path(output_dir)\n", " agg=scores.groupby(['method','budget'],as_index=False).balanced_accuracy.mean()\n", " fig,ax=plt.subplots(figsize=(10,5.6))\n", " for method in ARMS:\n", " rows=agg.query('method == @method').sort_values('budget')\n", " ax.plot(rows.budget,100*rows.balanced_accuracy,label=DISPLAY_LABELS[method],color=COLORS[method],marker=MARKERS[method],linewidth=2,linestyle='--' if method=='frozen' else '-')\n", " ax.set(xlabel='Original calibration-trial budget',ylabel='Mean balanced accuracy (%)',ylim=(0,100),xticks=[0,8,16,24,48],yticks=[0,25,50,75,100])\n", " ax.axhline(25,color='#aaaaaa',linestyle=':',linewidth=1);ax.grid(axis='y',alpha=.18)\n", " ax.set_title('Calibration curves: equal-weight mean across four development participants',pad=17)\n", " ax.legend(frameon=False,loc='upper left',ncol=2)\n", " fig.text(.5,.02,'Point estimates on the same later-session trials; participant-level variation is shown in the companion figure.',ha='center',fontsize=9,color='#555555')\n", " fig.tight_layout(rect=(0,.06,1,1));fig.savefig(output_dir/'mean_calibration_curve.png',bbox_inches='tight')\n", " return fig" ] }, { "cell_type": "code", "execution_count": 24, "id": "739e72e0", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [], "source": [ "def primary_comparison(intervals, output_dir):\n", " \"\"\"Expected columns: subject, contrast, estimate_pp, ci_low_pp, ci_high_pp.\"\"\"\n", " configure_style();output_dir=Path(output_dir)\n", " rows=intervals.query(\"contrast == 'blended_combined_minus_update_only' and budget == 24\").copy()\n", " wanted=['1','2','3','4','mean'];rows['key']=rows.subject.astype(str).str.replace('.0','',regex=False)\n", " rows=rows.set_index('key').loc[wanted]\n", " fig,ax=plt.subplots(figsize=(10,5.5));ys=np.arange(5)[::-1]\n", " for y,(key,row) in zip(ys,rows.iterrows()):\n", " color='#183a2c' if key=='mean' else '#276a9f'\n", " ax.hlines(y,row.ci_low_pp,row.ci_high_pp,color=color,lw=2)\n", " ax.plot(row.estimate_pp,y,'D' if key=='mean' else 'o',color=color,markersize=7)\n", " ax.annotate(f'{row.estimate_pp:+.2f} [{row.ci_low_pp:+.2f}, {row.ci_high_pp:+.2f}]',xy=(1.02,y),xycoords=('axes fraction','data'),va='center',fontsize=10,color=color)\n", " ax.axvline(0,color='#8b8b8b',linestyle='--',linewidth=1)\n", " ax.set(yticks=ys,yticklabels=['Participant 1','Participant 2','Participant 3','Participant 4','Four-person mean'],xlabel='Blended + update minus supervised update only (percentage points)',ylim=(-.6,4.6))\n", " low=min(rows.ci_low_pp.min(),0);high=max(rows.ci_high_pp.max(),0);margin=max(1,(high-low)*.08);ax.set_xlim(low-margin,high+margin)\n", " ax.grid(axis='x',alpha=.15);ax.set_title('Primary comparison at 24 original calibration trials',pad=18)\n", " fig.text(.5,.025,'95% paired run-bootstrap intervals, conditional on these four people, fitted models,\\ncalibration prefixes and one day-two session. These are not population or future-day intervals.',ha='center',fontsize=9,color='#555555')\n", " fig.subplots_adjust(left=.18,right=.68,bottom=.22,top=.87)\n", " fig.savefig(output_dir/'primary_comparison.png',bbox_inches='tight')\n", " return fig" ] }, { "cell_type": "code", "execution_count": 25, "id": "33f769d0", "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABkUAAARYCAYAAAC/P9Q8AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAViAAAFYgBxNdAoAABAABJREFUeJzs3XV0FNffBvBnVuLuhOAOQYNL0AaHQoDiVqCFKqW01F7gV6pUoRQKLVLcSqFFijsBWoK7exJCXFfm/WOzw26ymmywPJ9zcjI7986dO7OzszvznXuvIIqiCCIiIiIiIiIiIiIiouec7ElXgIiIiIiIiIiIiIiI6HFgUISIiIiIiIiIiIiIiEoEBkWIiIiIiIiIiIiIiKhEYFCEiIiIiIiIiIiIiIhKBAZFiIiIiIiIiIiIiIioRGBQhIiIiIiIiIiIiIiISgQGRYiIiIiIiIiIiIiIqERgUISIiIiIiIiIiIiIiEoEBkWIiIiIiIiIiIiIiKhEYFCEiIiIiIiIiIiIiIhKBAZFiIiIiIiIiIiIiIioRGBQhIiIiIiIiIiIiIiISgQGRYiIiIiIiIiIiIiIqERgUISI6DklCAIEQUCbNm2edFXoOXT9+nXpGBs+fLjJPMOHD5fyXL9+3WIZJfU4LQnb36ZNG4vHwbPEcFvo+VO+fHkIgoDy5csXqZwpU6ZIx8nu3bsdUjcq2Rx1bNLjd+XKFbz22muoWbMmPDw8pHODj4/Pk64aERFRiaZ40hUgInoczp07h8WLF2P//v24cOECUlJSIIoiPD09UaZMGVStWhURERGIjIxEo0aNIJfLC5Rx/Phx/PnnnwB0N8ae55uYj8Off/6JpUuX4t9//8X9+/fh7u6OcuXKoXv37hgzZgxCQ0NtLkuj0WDZsmVYsWIFTp48ifj4ePj4+KBy5cp48cUXMXr0aKsXn+np6di3bx/+++8//Pvvv7h69SoSExORmJgIuVwOf39/hIeHo3Pnzhg0aBD8/PyKuAfoeXP9+nUsXLgQAFCvXj28+OKLT7Q+RIZ4fD6/fvjhByQnJwPQBWSI6OkQExODDh06ICMj40lXhYiIiPJhUISInmvZ2dl488038euvv0IUxQLp+pvex48fx6pVqwAAY8aMwS+//FIg7/HjxzF16lTpNYMihZOUlIQBAwbgn3/+MZqfnZ2NxMREHDt2DD/88APmzp2Lfv36WS3vxo0b6Nu3L44ePWo0Pz4+HvHx8Th48CB++OEHLFmyBG3btjVbzvbt29GrVy+z6ZmZmbh16xY2b96MyZMnY+bMmRg0aJDV+lHJcf36dekcMWzYMN50pqcKj8/n1w8//IAbN24AYFCE6Gkybtw4KSDSuXNn9OrVCwEBARAEAUql8gnXjoiIqGRjUISInlsajQbdunXDjh07pHlt2rRBhw4dUL58eTg7OyM5ORkXLlzA4cOHcfDgQWg0Gmg0midY6+dbdnY2unXrhoMHDwIAAgMDMWrUKISHhyM1NRXr1q3D1q1bkZKSgoEDB8LV1RXdu3c3W96DBw8QFRWFixcvAgDKli2LUaNGoUqVKoiPj8eyZctw+PBh3L17F927d8euXbvQqFEji3WsXr06GjZsiOrVq6NMmTJwd3dHZmYmLly4gHXr1uHs2bNISkrCkCFDIAgCBg4c6Lgd9JxZuHCh9GQ6mWYqWEtERGToWe96sCS6c+cOYmNjAQDVqlXD33//DZmMvZcTERE9LRgUIaLn1ty5c6WAiI+PD9atW2exdUdSUhKWL1+OrKysx1TDkmf69OlSQKRq1arYtWuXUTdZr776Kr799lu8++670Gg0GDVqFC5dugQvLy+T5b3//vtSQKRFixbYtGmTUd433ngDb731FmbOnImMjAyMGDECJ06cMNk9WosWLXD37l2UKlXKbP0//fRTTJ06FVOnToUoinjzzTfRp08fODk5FWp/EBEREdHz59atW9J0vXr1GBAhIiJ6yvCbmYieW4ZPqH/++edWu7vy9fXFuHHjMGHChOKtWAmVmpqKr776Snq9ePFik+OGTJgwAZ07dwag6wLr+++/N1nepUuXpPfY2dkZy5YtKxA8EQQB3333HWrXrg0AOHPmDJYsWWKyvMDAQIsBEX15U6ZMkcpLTEzEgQMHLC5DRERERCVLdna2NO3i4vIEa0JERESmMChCRM+tc+fOSdNFGf9j4cKFEAQBI0aMkOZNnToVgiAU+Nu9e3eB5QRBsKkLofLly0MQBJQvX95q3n/++QfR0dEoXbo0nJ2dERoaio4dO2Lp0qVWu+M5duyYVC9b90tKSgo8PDwgCAICAwORk5Nj03KG1q9fL/Wr3LJlSzRu3NhsXsPA1LJly0zmWbFiBbRaLQCgb9++KFu2rMl8CoUCb731lvR66dKldtc9v1q1aknT9+/fL3J59srKysIvv/yCXr16oXz58nB3d4ezszNKly6NqKgofPbZZ7hy5YrJZdPS0rBq1SqMHTsWTZo0QUBAAJRKJTw9PVG5cmUMHDgQf//9t0O6dRo+fLh0rNna9UdycjK++OILNGzYEP7+/nBzc0O1atXw1ltvWS1j9+7d0vqGDx8OQPf+TJkyBREREQgICIBMJiswlkJiYiIWLVqEkSNHokGDBvD19YVCoYC3tzdq1KiBl19+Gfv27bO6XsMxaxYtWmTyHJH/XGDPZ/Ho0aMYO3YsatasCR8fH7i4uKBMmTLo2bMnFixYALVabXF5w3OSftyB+/fv45NPPkHt2rXh5eUFDw8PhIeHY9KkSXjw4IHVOhXWn3/+iR49eqBMmTJwdnZGqVKl0LNnT/z9998Wl7t+/XqB99gSe47Bc+fO4dVXX0WlSpXg6uqKgIAANG7cGF9//TXS0tLs2Dqdffv2YdCgQShbtixcXFwQEhKC1q1bY86cOVCpVADsO+9nZGRg5syZ6NKlC8qUKQMXFxd4e3ujVq1aeO2113Dq1CmTyxXl+LRVdnY2/vrrL7z99tto2bIlgoOD4eTkBHd3d5QvXx7R0dFYunSp1WMUePSZ0O8TrVaLJUuWoEOHDggJCZHOdf369ZNaHtoiJiYGQ4cORbly5eDi4oLg4GBERkZi9uzZyM3NLdR2F4VarcaOHTswadIktG3bFqVLl4aLiwtcXV1RunRpdO3aFXPmzLHYelV//OjHEwFg8n21dH65desW/u///g/NmzdHSEgInJycEBAQgKZNm2Ly5MlISEiwuB1TpkwpcPwcP34c48aNQ/Xq1eHl5QVBEPDDDz/Ys3sA6I75n376CVFRUQgNDYWLiwvc3NxQpkwZ1KtXDy+99BJ+/vlnq+cqlUqFhQsXIjo6GuXKlYObmxs8PDxQpUoVjBw5Evv377daF1EU8ccff6Bfv36oVKkS3N3d4eTkhJCQEISHh6NLly6YNm0aLly4YLaM7du3Y9iwYahevTo8PT2hVCoRFBSEGjVq4IUXXsDHH3+MY8eOmVzWnnPFhQsX8M4776BevXrw8/OTfidGRUVh5syZVltEm/ou1T/Y0rBhQ/j6+sLNzQ1Vq1bF66+/jps3b1qtky0Mz9f639O7du1C//79Ub58ebi4uCAoKAhRUVFYsmSJXb9Tzp07h4kTJyIiIgKBgYFwcnJCcHAwWrdujenTp1s9x5uq2549ezBs2DBUrlwZ7u7uEAQBf/75p/ReWTvnGl4z6ImiiLVr16J///6oUKEC3Nzc4OnpiSpVqmD48OHYvn17ofajuboCpt/vhIQETJ06FfXq1YOvry+8vb0RERGBGTNmGAV7AN1vqC+++AIRERHw9/eHu7s7ateujc8++8zqsSaKIg4dOoQpU6agY8eO0ufT2dkZISEh6NChA7755hukpKRY3e42bdoU+L7fu3cvBgwYIHWbHBAQgA4dOth9/OzZswdjx45F7dq14e/vD6VSCR8fH0REROC1117D1q1bpWsScw4fPozXX38dtWvXNvpcdurUCXPmzCnUtRURERWRSET0nHJ1dRUBiADEXbt2FbqcBQsWSOVY+zNcj+FyCxYssLqecuXKiQDEcuXKmc2jUqnEIUOGWKxDVFSUmJaWJr1u3bp1gXKaNm0qpZ89e9Zq3WbMmCHlf++996zmN6V///5SGV988YXFvCqVSnR3d5fyX7x40eI2LF++3GJ5d+/elfIqFAoxMzOzUNugFxERIZW3c+dOs/nMHRtF8ddff4nBwcFWj0UfH58Cy+bk5IguLi42Hcvt2rUTHzx4YLYe165dk/IOGzbMZJ5hw4ZJea5du2axjNatW4unTp0Sy5cvb7ZObm5u4ooVK8zWadeuXUZ12r59u+jv71+gnJ49e0rLXLlyRVQoFDbtk/79+4sZGRkW12vtL/+5wNLnVC8nJ0ccOXKk1bKrVatm8fNseE6aPHmyuH37djEgIMBseaVKlbLp/GBN69atpTIvXbokDhgwwOp+zs3NNVmWLcedIWvHoN7MmTNFpVJptk6VKlUSz5w5Y7Qt5mi1WvHNN9+0uI2NGzcW7927Z9N5XxRFccOGDWJQUJDFMgVBECdOnChqNBqjZYtyfNqqQoUKNpVft25di++DKD76TJQrV058+PCh2K5dO4tlfvfdd1br9/7774uCIJgtIyIiQrxz547N74c1kydPlso2d+5v27atTfusXLly4rFjx0yWoa+vtT9T5xetViv+73//E52dnS0u6+7ubvG8a7itCxYsEKdPn27ynPr999/btQ9PnDghlilTxqbts/Sb4tChQzYdn4MHDxazsrJMlpGUlGT02bf017FjxwLLZ2dni/369bNp+WrVqpmsgy3HplarFT/88ENRLpdbXEdoaKi4b98+s+Xk/y6NjY21uA89PT3FvXv3mi3PVobn6127donvvvuuxe1o27atmJKSYrHM3Nxc8bXXXhNlMpnFsgIDAy3+Tstft7fffttkOevWrbP5c5l/fXfu3BGbNGlidblu3bpZ3G576iqKBd/vQ4cOiaVKlTK7/pYtW4ppaWmiKIri0aNHxdDQULN5GzRoICYlJZmt64gRI2zaV/7+/uK2bdssvteGn9GrV6+K48ePt1hmdHS0qFKpLJZ57949sX379jbV8c8//zRZRmpqqti3b1+ry1esWFE8efKkxfoQEZFjcUwRInpuValSBSdPngQAfP/992jdujUEQbC7nHbt2mHdunXYuXMnZs6cCQB46aWX0L9//wJ5w8PDi1ZpK0aOHInFixcDAORyOfr164f27dvD3d0d58+fx4IFC7B161ajVi2mvPbaa4iJiQEAzJkzBz/++KPF/L/88gsA3ROor7zySqHqrn8vAFhsJQLoWnfUr19fenrz5MmTqFKlipQuiiJOnz5tc3mlSpVCWFgYbt++DbVajbNnzyIiIqIwm4EffvgB//33HwAgNDQUzZs3L1Q5hTF//nyMGjVKerqtWrVq6NOnD6pWrQoXFxfExcXhv//+w6ZNm0w++azVapGdnY3AwEC0a9cOderUQZkyZeDm5oa0tDScOXMGq1evxo0bN7Bz50707t0bO3fuNDkGi6OlpKSgZ8+euH79Olq2bIm+ffsiJCQEt2/fxrJly/Dff/8hMzMTgwYNgre3Nzp16mSxvMuXLyM6Ohqpqal48cUXERUVBX9/f9y+fdvoycXc3Fyo1WqEhYWhXbt2CA8PR6lSpeDq6ork5GQcP34cq1atQnx8PFasWAG5XF6gC7bw8HCsW7cOp0+fxieffAIAaNu2Ld58880C9WrQoIFd+0UURURHR0stKJRKJQYOHIjIyEi4ubnh3LlzWLhwIW7evIkLFy6gefPmOHr0KCpXrmyx3BMnTuCbb75Bbm4uBg0ahNatW8Pb2xvXrl3DvHnzcOXKFdy7dw/9+vVDbGwsFArH/GScNGkS1q5di5CQEIwYMQK1a9dGdnY2du3ahWXLlkGj0WDFihWQyWQOadVli/nz5+ONN96QXkdGRiI6OhohISG4c+cOVq5cicOHD6Nnz57w8PCwWt7EiRMxY8YMALpzZs+ePdGpUyf4+vri+vXrWLp0KY4cOYJ+/frZ1HLi999/x4gRI6DVaiGTydC5c2e0b98epUuXRm5uLmJjY/H777/jwYMHmD59OrKzs6X1A8V7fOplZmbCy8sLbdu2Rf369VG+fHl4eHggMzMTFy9exB9//IGzZ8/ixIkT6NSpE/7991+r+1KtViM6Ohq7du1Cw4YN0bdvX5QrVw5paWn466+/sGHDBgDAu+++iyZNmpg9F//f//2fUdeN3bp1Q9euXeHr64tr165h8eLF+O+//9CnTx+pBc/jkJmZCVdXV0RGRqJhw4aoWLEivLy8kJubi6tXr2LDhg04evQobty4gY4dO+L48eMFupycO3cuMjMzMWbMGKlFx7p16wqsKyAgoMC80aNH47fffgMAeHh4oE+fPmjatCn8/f2RnJyMvXv3YuXKlcjIyED//v2hUCgQHR1tcZtWr16NTZs2wd3dHaNGjUKTJk3g7OyM8+fPo0yZMjbvm6ysLPTo0UMak6F27dqIjo5G5cqV4eLigtTUVFy6dAmHDh2y2IXl9u3b0a1bN+kJ7NatW6Nz584oV64cRFHE2bNn8fvvv+PmzZtYsmQJUlNT8eeffxb4rfjKK69gz549AIDg4GAMHDhQal2XlZWFGzdu4L///pPGscvv//7v/7Bq1SoAgJeXFwYOHIiIiAj4+voiOzsb9+7dw7Fjx2xqBWDJG2+8gVmzZgHQnXt69eqFqKgo+Pj44OrVq1i8eDHOnTuHu3fvokOHDtixYwdatGhhsczbt2+jc+fOiIuLQ8+ePREVFYXAwEDcuXMHixYtwvHjx5GWloZ+/frh3Llz8PHxKdI26P30009Yu3YtvLy8MHLkSERERECr1eLw4cNYsGABsrKysGvXLnTr1g27du0y+TtFo9Gge/fu+OeffwAA/v7+6NevHyIiIuDt7Y2EhARs27YN69evR0JCAjp27Ig9e/agadOmFus2ffp0bNq0Cf7+/hg2bBjq1q0LmUyGkydPwt/fX/pcWjvnGl4zJCYmonnz5lKrr4CAAIwYMQJ169aFRqPBkSNHsHDhQmRkZODvv/9Gu3btcODAATg7OxeprvndunULPXr0QFJSEgYPHoy2bdvC3d0dp06dwqxZs5CcnIz9+/djwoQJ+OCDD9ChQwfk5ORg1KhRaNGiBVxdXXH8+HH8/PPPSE1NxbFjxzBhwgTpXJNfZmYmlEolmjVrhqZNm6Jy5crw8fGBRqPBzZs38c8//2Dnzp1ITExEjx49EBMTgzp16ljcZkD3mVuyZAnCwsIwdOhQ1KxZEwBw4MABzJ8/Hzk5OVi7di2++eYbTJo0yWQZt2/fRtOmTXHnzh0AgLu7O/r27YtmzZrBz88PaWlpOHfuHLZt24bjx4+bbHmSnp6OVq1a4cSJEwCAsLAw9OvXD7Vr14a7uzvu3buHv/76C9u3b8fVq1cRGRmJf//9F5UqVbK6jURE5ABPNCRDRFSMvv76a6MncJo0aSIuWLBAvHXrVqHKy/+UtT35HdFSZN26dVJ5Hh4eJp/KS09PF7t06WK03aaeEM3OzhYDAwNFQNeiwNTT73p79uyRyurUqZPV7TBFq9UaPYV99epVq8sYtoiZNm2aUdqtW7ekNLlcbvapckOtWrWSllmyZInFvDk5OeK6deukv6VLl4r/+9//xIYNGxo9Gbl9+3aL5Ri+D0VtKXLs2DHRyclJBHRPhH/++eeiWq02mVelUklPABpSq9Xi33//XeBJckM5OTnimDFjpHovW7bMZD5HtxTR/3355ZcF8mk0GnHChAlSntDQUOkpRUP5n4h3dXUV//nnH7PbKoqimJiYKO7Zs8dinrS0NLFbt25SuQcPHjSZL//Tlraw9DkVRV0LBn2ewMBAMTY2tkCe/J/7xo0bmywrf6u3UqVKicePHze5vfXq1ZPy/fHHHzZtizn5n7Bu1qyZySdH9+/fL3p4eEj51q5dWyCPo1uK3Lp1y2idplodaLVaccqUKQWOVVMOHjwotUhwcnIS//rrrwJ51Gq1+MYbbxiVZe68f/r0aal1V2hoqHjkyBGT+R4+fChGRkZK5ZlqwVaY49NWGzduFHNycsymazQacerUqdL6P//8c7N58+/nr7/+2mS+Tz/9VMrTo0cPk3n+/fdf6al5hUIhrlq1qkAelUolvvzyyza9H7aypaXI1q1bTZ7HDM2fP186nsaMGWM2n+GT6bb47bffpPytWrUS7927ZzLfqVOnpCfGvb29TX5uDbcVgFi5cmWrrYGsWbt2rVRenz59zH7XiaIoxsfHi6dPny4w//79+1JLOE9PT3Hz5s0ml8/KyjJ6inv+/PkFyte3MqhUqZKYmJhoti5ZWVnioUOHjOap1WrRx8dHBCD6+vpa3DdqtdpsiwtrvxH/+usvaRvc3NzErVu3FsiTm5trdKyXLVvWZOuY/N+lHh4e4o4dOwrkU6lUYseOHS2eP+1heL4GIFapUkW8efNmgXznzp0zasnw7bffmizvk08+kfJER0ebbV2xe/du6XugUqVKJo+3/HVr3Lixxda0omjfOTc6OlrK26RJE5NlX7lyxag17cSJE02WZW9d87/fPj4+YkxMTIF8ht9HSqVSrFu3rhgWFiaeP3++QN7jx49Lv1kVCoUYFxdnct179uyxuh83bdokrTcqKspsvvy/NaKjo022DN+9e7f0vRAYGGjyu0ur1YqNGzeWymrZsqXZ86Qo6lq2XbhwocB8w2uZ119/XczOzja5/IoVK6Q6RUZGml0PERE5FoMiRPTcys7ONrpJZPgXFBQkdurUSfzoo4/ETZs22dSd0pMOihg2qZ83b57ZclJSUowuFs3dbP3www+lPL/99pvZ8gy7ulm/fr3V7TAlNTXVaP+npqZaXcaw2fu7775rlHby5Ekpzc/Pz6Y69OrVS1rmp59+spg3ISHB5HEDQHR2dhZ79+5t8iIwP8PlihoUMbwpn39/OJpKpZKOR3MXoMURFOndu7fZOmm1WrFFixZS3tmzZxfIk//C/ptvvrFpe22RnJwsurm5Wbw56eigiEqlEsPCwqQ8GzduNFtOamqqUV5TwaD8QRFLQb3NmzdL+UaNGmXTtphjeKPCw8NDvHPnjtm8c+fONboxlJ+jgyLvv/++lD5o0CCLZXXt2tVo/5nSu3dvKf3TTz81W5ZGoxEbNWok5TV33u/Tp48IQJTJZOLRo0ct1i8+Pl709PQUAYhdunQpkF6cQRFb6T/DVatWNZvHcB8PHjzYbD61Wi2WLl1aBHQBUFPdoLz00ktSWR999JHZslQqlVi3bl2r74etbAmK2GrgwIEioOvGylxXL/YERXJycqT9FhYWZrX7oS1btkhlmwpQ5Q+KWDtObfHFF1/YdN6zxLDrJWuB3ezsbGkf1qxZ0yjt0KFDUjnmbkJbcu/ePWn5vn372r28nrXfiIZdis6aNctsOWq1WmzQoIGU95dffimQJ/936a+//mq2vHPnzkn5OnToYPd2GTI8X8tkMpMPAeht3bpVylu6dOkCn434+HjpO7tBgwZWu0maM2eOVJ6p4Klh3VxdXU0Ga/Kz9Zx77tw5Kfjp5eUl3r1712zeo0ePSkE6d3d3k0E6e+ua//3+/fffzebNH0DevXu32bzDhw+X8i1evNhiHawxvGYx9xvC8LdGlSpVzHaHJ4rG3w379+8vkL5mzRopvWLFijZdt+R36tQp6X01F7g3NGnSJGmd5h6AICIix+JA60T03HJ2dsbWrVvx0UcfwdPT0ygtPj4eW7ZswWeffYYuXbogODgYY8eOlZpIP21u3bqFw4cPAwCCgoIsDjDs5eWF1157zWqZr7zyitTdwOzZs03mSUhIwNq1awEAZcqUQdeuXe2suU7+ASxdXV2tLmOYJzU11Wx5tpRlrTx71KhRAx06dECpUqWs5hV1Dx9AFEWbB7U3JTExERs3bgQAeHp6St0xFBeFQiF1HxETE+OQQddt8d5775lNEwTBKH316tUWy3J1dcWYMWMcVjdvb2/Url0bAHDo0CGHlWtJTEwMbt++DQCoW7cuunTpYjavp6cn3nrrLem1tf1Tt25dtG/f3mx627ZtpS6zDLuqK6rBgwcX6ALI0IgRIxAUFARANyipvvuc4mK4n95//32LeT/44AOL6bm5udLn1NnZGa+//rrZvDKZDBMmTLBYXmpqqtQVUocOHdCwYUOL+QMDA6Vz9K5du57KQVtbtmwJALh48SIePnxoNb+lfSSXy6WBjLOysnDlyhWjdJVKhfXr1wMAnJycMH78eLNlKRQKTJw40Wp9noRWrVoB0A06btgNZWFt27ZN+q0zbtw4eHl5WczfsWNHhISEAAA2b95sMW/Lli2tHqe2cHd3l6b13VXaQxRFLFq0CICum8levXpZzO/s7IwBAwYAAM6ePWs0cLhhXcwNgm6Jm5ub1B3XqVOnTHZtWVS3b9+WukQNCAjA6NGjzeaVy+VG5zJr3xUBAQEYNmyY2fTq1asjLCwMgGO/K6KiolCvXj2z6S+88ALq168PALhz506B7+VVq1YhMzMTgK6LPWtdQA4aNEjKY+04j46Otqs7OGvWrFkj/c4aMWKExd+XDRs2ROfOnQHozgmOrmtgYCAGDhxoNl1/PgKA+vXro3Xr1mbzGqadOXPG5jpYW6/+WLdk3LhxcHFxMZv+wgsvSNOmjlv9+QMAPvnkkwLXkbZYtGiR9L6a66LLkOG1nbX3lYiIHINjihDRc83Z2RnTpk3D+++/jw0bNmD79u2IiYnBxYsXodVqpXxpaWmYM2cOVq5cidWrV1u8Wfgk6AMigG6ME2sXd1FRUfj4448t5ilbtiy6d++OP//8E//++y/+/fffAjcz5s+fL13Ajxkz5rGMLfE0CAgIkC5kRFFEamoqzpw5g5UrV2L27NkYN24cvvvuO/zxxx/SjfLitG/fPqk+7du3t3oTy5r4+HgsXboU27Ztw9mzZ5GYmIiMjAyTwY/U1FSkpaUVeZ3WeHl5WR0bpn379pDJZNBqtThy5AhEUTQ7TlC9evXsuoi9ceMGlixZgt27d+PcuXNISkqSbqjkpw9UFDfDC/+OHTtazd+5c2fpxq61mwbWxsJxdnZGQEAA7t+/j6SkJBtqa5sOHTpYTFcoFGjTpo3U//7hw4cdevPJUEJCAq5evQoACAkJsfpZbtasGTw8PJCenm4y/fjx41IgIiIiwmrf+tb2xf79+6HRaADognJ//vmnxfwApPN1VlYWrl27hurVq1tdxlFSU1OxYsUKbNq0CadPn0Z8fDwyMjKMvmsN3b59G35+fmbLc3d3R926dS2uU38zFkCB4/TEiRPIzs4GoBsrxVT/+YaioqIspheHnJwcrFmzBn/99RdOnDiBe/fuIT09XXrf87t9+3ahx33R04+NAegCR7YcV15eXrh//z7Onj1rMZ/hjcuiiIqKks71//vf//DgwQMMHToUDRo0sGlsuHPnzkljrISEhNi0jcnJydL02bNnUbZsWQBArVq1UKZMGdy6dQs7duxA9+7d8frrr6NNmzZWx3MAdPuuWbNmOHjwIM6fP4/27dtj/PjxiIqKsmmMIlsYnu/btWsHpVJpMX/Hjh2l/Wvtu6JRo0ZWf3Pqx2x7nN8VgO7GdmxsLADdd4Xh8Wd4nKekpNh0DHh4eCA5OfmxHed6hfmu1wfgY2JiMGjQILN57a1ro0aNLP7W1wdIAd13oiWGwR1Lx4ZWq8WGDRuwbt06xMbG4vbt20hLSzM75pYtv8Gs/cax9N0BAHv37gXwaFywwtAfg4Ig4M6dO1aPQcMxrawdg0RE5BgMihBRieDp6YlBgwZJFw7p6ek4fvw49u3bh5UrV0oD4CUlJeHFF1/EiRMnULFixSdZZSOGLVgMBxw3p2rVqjaV+9prr0k/0ufMmYNff/1VShNFEXPnzgWgG9x51KhRdtTYWP6b01lZWVZvWBsOhp3/hrzhsob5ClueJYIgwNvbG82bN0fz5s0xYMAAdOjQAZcvX0bbtm1x+vRpo4vE4mD4tLx+sMjCWrBgAd56660CrXcsSUlJKfagSKVKlaze7HJ3d0doaChu376N9PR0i8Eae26kf/7555g6darNT/CmpKTYXHZRGH7uq1WrZjW/YZ67d+9azGtq4OX89Df89DeWHcHe85e17SgKw7JtqZdMJkPlypVx/Phxq+XZMkiqv78/fHx8jG7GGrp+/bo0vXr1aqtPdOdnS0sMR9m4cSNGjhyJ+Ph4m5ex9jny8/Ozek4wvCmd/zi193szMDDQ4vvhaEeOHMGAAQOkwJwtHHHuMTyuJk+ebNey1o4pS+ddazcEW7ZsKZ2XqlWrhqlTp+KTTz6BWq3GjBkzMGPGDPj6+qJp06Zo3rw5OnTogCZNmpg8Rgy3cc+ePUY3yG1huJ0ymQy//fYbevbsiaysLPz999/4+++/4ezsjIYNG6JZs2Zo27Yt2rdvbzZIMnv2bLRr1w6JiYnYv38/9u/fD4VCgXr16qFZs2aIjIxEp06dCh0ksfe7wtPTE6VKlcKdO3ek71Jzv8ns+a5wZOu0on5XGB4DY8eOtWvdRTnOC6M4v+vtrau14LHhMW5PXnO/Iy5fvow+ffpI12G2sOU8aO24tVS3tLQ0aR2lSpWCr6+vzXUzpD8GRVFE37597Vr2cX5/ExGVZAyKEFGJ5OHhgZYtW6Jly5b44IMPsGDBAowePRoajQbp6en46quv8MsvvzzpakoMb2AbduVgji15AN2TeNWrV8f58+exfPlyfPvtt/D29gYAbN26VbpZ06tXryLd+Pfw8IBCoZCe+nrw4IHVoMiDBw+k6fxPXBu+TklJgVqttvoko6Xy7NG0aVNMnDgRU6ZMQWJiIn788Ud88cUXhS7PFobdfRWmCb/eH3/8gZEjR0qvGzZsiDZt2qBixYrw9fWFs7OzdINpxowZ2LVrFwCYfWrZkWw9Zg3zWQqK2Nqt2nfffYePPvoIgC4AFhkZiRYtWqBcuXLw9vY2unD++OOPcebMGbNPvjuavZ97hUIBZ2dn5OTkWO0iTiZ7Mj2o2nv+sid4Zy9796+1fIYtSOwpz9xN+KLenC+ObnpMOXz4MHr16iU95VqzZk106NABlStXRkBAAJydnaXjbcWKFVi5ciUA6+eVoh6jhX1/H0dQ5MaNG+jQoYNUx/Lly6NTp06oVq0agoKC4OLiIm3/zp07MXPmTACOORcXZfsMn2Q2xdJ511oXVrt27TLqZvLjjz9GkyZN8OWXX2L37t3QarVISkrC5s2bsXnzZnzyySeoVKkSPv30U6nrKz1Hf3b0LRKmTZuGP/74A5mZmcjJycGBAwdw4MABfPPNN/Dx8cHbb7+NDz74AE5OTkbL16lTBydPnsS0adOwfPlyJCcnQ61WS610Z86cCVdXV4wePRrTpk2z+3u+MMe6YQAmNTXV7Dqf1e+KohwD1s6dtv6+sJW971/+984Se+tqz/td1GMjNTUVbdu2lVp+BAcHo3PnzggPD0dwcDBcXV2lViunT5+Wuo615TxYlLo56jd3cR6DRETkGAyKEBFB14dvbGysdONh69atj70Oln7kG/4oz8jIsFqWLXn0xo0bhzfffBOZmZlYtGgR3nzzTQDG44zY+5RdfoIgoGrVqlJz8GvXrqFChQoWl7l27Zo0nb8LmLCwMKkbG41Gg5s3b1pt2WOpPHt16dIFU6ZMAQApcFCcDG/8F+Umsb5PY7lcjlWrVqF3795m8y5durTQ6ykMW49Zw3xFuVgFdE8H6t9Hd3d3bN261WKXC5999lmR1mcvez/3arVaelK3uFv2FJa956+ivseOPK9ay2d4o8rR5f3vf/8r9rGECuuTTz6Rbpb/8MMPRmPb5HfgwIHHVS2Hv7+O9Nlnn0nn8gkTJuDrr782exPP0WOdGR5Xe/fudXhXQI70wgsv4IUXXkBSUhIOHDiAw4cPY//+/Th48CByc3Nx5coVDBw4EJcvXzb6fBhu48iRI/Hbb78VuS7VqlXD4sWLMW/ePBw5cgSHDx/GwYMHsWvXLqSkpCA5ORlTpkzBgQMHsGXLlgLvZ2hoKH7++WfMmDEDx44dQ0xMjLR8fHw8srKyMGPGDOzevRsHDx60ObgBFO5YNwziPo3fF0X9rjA8Bm7evFls3TA6gr3v39P+3tnqp59+kgIi/fv3x8KFC822trLWJZwjOeo3t4eHB5KSkiCXy6FSqWzq+o+IiB4vDrRORJTHsB9fR3TZYvjD3toTP1qtFomJiWbTS5cuLU1funTJ6rovXrxoQw11hg0bJl086lvH3L59G3///TcA3cDiRRkkXK9OnTrS9JEjRyzmVavVUj/R+ZcFdEGW8PBwm8u7d++edOEll8uL3AWV4QWsI/vQNkfftzlQ+MEqr127Jh07PXr0sBgQ0ed/nK5cuWJ1QPeMjAzps+nh4VHkG+aHDh2SLnjHjBljtQ9qw+44HgfDz/2FCxes5jfMY2kw8yfp8uXLVvMYnuPyb4c951UA0rgCpth7XhVFscBg3oYM62opn15iYqLFJ0kN+zw3HPj5aaJSqaTAcP369S0GRIDHe16x9/1NSEh4bF1n/fPPPwB0XXZ99dVXFp9qdvQ+e1LHlSiKFv8s/c7w9fVFt27d8Omnn0pBhP/9739S+qeffor79+9Lr4tzG11cXBAZGYmJEydi3bp1ePDgAVauXCm1st22bZvFrsIUCgUaN26MN998EytWrMD9+/exefNm6ab9yZMn7W6pbO93RVpaGu7duwfAMd+lxaGo3xXPwvlT73n8rreF/jyoUCgwa9Ysi2P0PM7vDk9PT6lF+b179wr9O19/DGo0mmLtCpSIiAqPQREiojyG3UKYukA0vGlh7eYtAKMBZA3HhDDl+PHjFsfGaNq0qTS9a9cus4MP6m3bts1q/fS8vLwwZMgQALqB/fbs2YN58+ZJT1i/+uqrNpdlSZcuXaTpTZs2Wcy7Z88e6Wm5KlWqmOxb2p7yDNPbtm1b5K4PDC/EAwMDi1SWLVq2bCk9YbZz506r3SWYYnjDyNqYM3fu3DE7bkJxSU1NtRrc2rlzp9R1lbm+5O1hzz45fPiwURdspth7jrDG8HOvv3lgyZYtW0wu+zSxdm5Sq9VGra/yb4dh397WzqsqlQpHjx41mx4QECCN/XH//n2cPn3aYnmGQTRT6tWrJ93U+e+//6zeYN++fbvF9MjISOmY2rp1a5G7TnL08QnouiXUfx9Z+wxlZWU9lpZ1enXr1oWLiwsA4NixY1b7aLfne7Oo9OeeihUrWhzUWBRFq99vgH3vrWHwwZayn0be3t745JNPpOC+SqUyGqy6bt260rni4MGDxToOlEKhQL9+/fDpp59K8/SDNNtCEAR06tQJs2bNKtTygPF5cufOnVa7Odu6dav0Xfqsflfkz5N/O56l4/x5/K63hf486O/vb3TNZIp+YPnHJTIyEoDufGptPCRznqVjkIiopGJQhIieW4Y3PG2xbt06abp27doF0g2b4hs2XTenVq1a0rS1i7uvv/7aYnpYWBiaNGkCAIiLi8Pvv/9uNm9aWhp+/vlnq/UzNG7cOGl65syZUlcTbm5uGDZsmF1lmdOjRw+pO4h9+/ZZvAH+7bffStMDBw40meell16SbgStWrXK7A1StVqNH3/8UXo9aNAgu+uen+H+bdmyZZHLs8bf3x/dunUDoHt/DW++2MqwKw5rLYkmT578WMYRye+bb76xmD59+nRpuk+fPkVen637RBRFfPzxx1bLs/ccYU3Tpk2lp4dPnDhh8aI6PT3d6Di3d1DPx2Xp0qXSE8qmLFq0SBqsu0mTJkZP+wK6p7T1gYyjR49aDDz89ttvVm+EGx5H1s7DX331lcV0JycndO3aFYBuwOGffvrJbF6tVmt0njMlICBA+tzfvHnTYnm2cPTxCdh3Xvn2228f6+CxSqUSPXv2BKBrVfTDDz+YzavRaKyefxxJv98uX75sMYixdOlSnDt3zmp59ry3nTp1ksYIW7lyJf777z9bqvxUMuw20/BhEblcLj3skZmZKXWT+CTq8jiWDwsLQ7NmzQDoApXz5s0zm1er1eLLL7+UXj+t3xXbtm3DyZMnzabv2LEDx44dA6BraZE/ONC/f38pKDpr1qynurVInz59pIc8Fi5caPE7MjY2Fps3bwagO48YPiD0rNGfB+Pj4y0GLvft22dTsMiRDK99pk2bVqhutIYPHy5Nf/HFF4+tJSIREdmOQREiem41atQIQ4YMwb59+yzedNBoNJg+fbpRoMHwh6ye4QWrLTcRwsLCpG6fDh8+jMWLFxfII4oiPv/8c2ngWUs++OADaXr8+PE4ePBggTxZWVkYOHCg3X2Qh4eHo3Xr1gCAtWvXSssPHDhQ6hKiqLy9vTFx4kTp9dChQ002J//222+lC76AgACMHz/eZHlVq1bF0KFDAehuQA4cOLBACwpRFDFhwgScOnUKgK4rsMGDB5ss76233rJ60ZyRkYFXX31VekrPxcUFo0ePNptfEATpb/fu3RbLtmbq1KlSn8rffvstvvjiC7MDfms0GmzYsMFoXo0aNaQWUH/99ZfJJ1G1Wi0+/fRTh/S/Xhhr1qwxeaNYFEVMmjQJ+/btA6DrLsLc+2iPRo0aSTcifv31V5PdVuTm5mLs2LFWn+oHjM8RsbGxRX4aX6FQ4P3335deDx8+HCdOnCiQLzMzEwMGDJACg40bN0ZUVFSR1l1c0tLS0LdvX5M3QGJiYvDOO+9Irw233ZBh4OHdd981mWfr1q1m0wy9/vrr0g3lxYsXS+NK5Tdt2rQCnylT3n33XemYmjZtmsmnWzUaDd5++22LrVj0PvvsM+nG3rvvvotZs2ZZPK5ycnKwcuVKkwEURx+fgK6loX6MptjYWCxbtsxkvt9+++2x3JjOb+LEiVJLjC+//BJ//PFHgTxqtRrjxo0z6rKxuOkfckhMTMQXX3xhMs+mTZtsbqlpz+8TNzc3TJs2DYDuWOzatavVFjyJiYn49ttvbToPOsKMGTOwYsUKaYwkU+Lj47Fq1Srpdb169YzSP/zwQwQEBADQjXXz8ccfW2xBodFosGnTpgIPHfzzzz/49ttvLXZxqlKpMGfOHOl1/fr1penY2FhMnjxZ6sLTFFEUjT6zhsvb6qOPPpKmJ06ciB07dhTIo1arMXbsWPz7778AdF1z6n9HPW00Gg369etn8vfspUuXjG5aT5gwAQqF8VCpoaGh0ndASkoKOnToYPL709CdO3cwefJki8GY4lC9enVER0cD0NW1d+/eJgPI169fR58+faSHVsaNG2fUevJZoz8PiqKI9957z+R30tGjR9GnTx+HtW60Va9evdC4cWMAwNWrV9G1a1fExcWZzX/69OkC3TQ2aNBA+q167do1dOrUyWo3rJcuXcL48eOlh0OIiKh4caB1InpuqVQqLFmyBEuWLEFoaCgiIyPRoEEDhISEwM3NDSkpKThz5gzWr19v1P97t27dTF4k1q5dG6VKlcK9e/ewZ88ejBo1ClFRUUYD8jVu3NioCfgHH3yAAQMGANDd0NyyZQuioqLg6uqKa9euYeXKlYiNjUXbtm1x8eJFi8GMnj17YsiQIVi8eDFSU1MRGRmJ/v37o127dnB3d8f58+exYMEC3LhxA3379sXq1avt2l+vvfYa9uzZYzSvqAOs5/fee+9hy5YtiImJwYULF1CvXj2MHj0a4eHhSE1Nxbp166SnweRyOebNm2cxKPP1119j//79uHz5Mvbv3486depg1KhRqFy5MhISErBs2TKpSw03NzcsWLCgwIWz3owZMzBz5kw0btwYzZo1Q/Xq1aWLzYSEBMTGxmL9+vVSF0qCIOCnn36SnlovbvXr18esWbPwyiuvQBRFfPjhh1i0aBH69OmDatWqwcXFBfHx8YiNjcXGjRuRlZVl9FSaUqnEG2+8gc8//xxqtRrt27fHoEGD0KxZM/j4+ODKlStYsWIFTp06hdDQUISHh2Pr1q2PZdsA3Q2ttLQ0vPvuu1i/fj369u2L4OBg3LlzB8uXL5duIMvlcvz2229GT0YXVqlSpTBgwAAsW7YMaWlpqF+/PkaOHIn69evDzc0N586dw9KlS3H16lXUrl0bzs7O0s0kU3x8fNCoUSMcPXoU165dQ3R0NKKjo+Hn5yfdKK9du7ZR/+HWjBs3Dlu2bMHff/+NhIQENG7cGAMHDkTr1q3h6upq9LnX12Hp0qVF2zHFKDo6GmvXrkWNGjUwcuRI1KpVCzk5Odi9ezeWLl0qPSE9cOBA9OrVy2QZb731Fn799VdkZmbit99+w8WLF9G/f38EBQUhLi4OW7ZswcaNGxESEoKmTZuavDmoFxYWhh9//BEvv/wyAODNN9/E2rVr0adPHwQFBeHu3btYuXIlYmJiULlyZXh4eFjsWq5Zs2YYP348vvvuO+Tk5KB79+548cUX0alTJ/j4+OD69etYtmwZTpw4gVatWuHq1au4c+eO2TElwsPDsWjRIgwcOBBqtRqvv/46fvzxR7z44ouoVasWPDw8kJ6ejps3b+LYsWPYsWMH0tLSpO0xVBzHJwC88847GDNmDABg8ODBWLt2Ldq1a4fAwEDcunULf/zxBw4ePAhPT0907NgRa9assav8ooiIiMCHH36ITz/9FCqVCtHR0ejRowe6dOkCX19fXLt2DYsXL8aZM2fQvHlzXL9+/bH0/T5+/HgpuP7RRx9h586d6NatG0JDQxEXF4dNmzZhy5YtUCgU0ve+JVFRUVi/fj0A4OWXX8Zbb72FihUrSt93fn5+0g0+fZ4TJ05g5syZiIuLQ7t27dCqVSt07NgR5cuXh5OTE1JSUnDp0iUcOXIE+/fvh1qttloPRzl27BgWLVoEDw8PdOjQAREREShbtixcXV2RmJiI2NhYrFq1SvqOGzBgACpXrmxURnBwMP744w906tQJmZmZ+Oyzz7Bw4UJER0ejbt268Pb2RmZmJu7cuYMTJ05g27ZtSExMRPv27Y0Gbb937x7effddTJo0CS1btkSTJk1QtWpVeHl5ITU1FZcvX8aKFSukMQ8qV66M/v37S8unpKTgf//7Hz799FM0btwYzZs3R40aNeDj44OMjAzcuHEDa9askbrv8/PzM2q9a6uuXbvitddew6xZs5CZmYkXXngB0dHRiIqKgre3N65evYrFixfj7NmzAHTjMy1btkwKuj5t9N8VtWrVwsiRIxEREQFRFBETE4MFCxYgMzMTANCqVSu8+eabJsuYOnUqTp06hfXr1+PSpUto0KABoqKi0K5dO4SFhUEulyMpKQnnz5/HoUOHcOTIEYiiiBdeeOFxbioAYPbs2Th69Chu3LiBmJgYVK9eHSNGjEDdunWh0Whw5MgRLFy4UGoJFhERUahWw0+TN954A7/++ivUajXmzp2LY8eO4aWXXkKZMmWQnJyMHTt2YO3atdBqtRgxYgQWLFjw2OomCAJWr16Npk2b4t69e9i3bx8qVaqEfv36oWnTpvDz80N6ejouXLiAHTt24OjRo1i3bl2B7n5/+eUXXLp0CYcPH8bhw4dRtWpV9OjRA61atUKpUqWkMSXPnj2L/fv3SwG5CRMmPLZtJSIq0UQioudUx44dRUEQRAA2/SmVSnH8+PFibm6u2TIXLVpkscxdu3YVWGb8+PEW19u+fXvx4cOHYrly5UQAYrly5cyuX6VSiYMHD7ZYXseOHcW0tDTpdevWrW3aXyqVSgwNDZWWa9y4sU3L2SsxMVF84YUXLG6Dl5eXuGzZMpvKu3r1qhgREWGxvJCQEHH79u0Wy7H1OAEgli1bVtywYYPVulk7Ngpj7dq1or+/v9U6+vr6Flg2NzdX7NGjh8XlKlasKMbGxorDhg2T5l27dq1AWdeuXZPShw0bZrKu9pTRunVr8fTp02KFChXM1s3NzU1cvny52X2za9cuq3XKLyUlRWzatKnFfVK/fn3x+vXrYuvWraV55uzYsUNUKpVmy1qwYIFRfls+pzk5OeKIESOsvudVq1YVz5w5Y7acBQsWSHknT55sdd/Yck6yheF+u3Tpkjhw4ECL2/HSSy9ZPA+LoiiuXr1adHJyMltGhQoVxOPHj1s9BvVmzJhh8X2rXLmyePbsWZuOAa1WK77++usWt7FRo0bivXv3xNKlS4sAxDp16ljc3j179lj8bBj+CYIg/t///Z/Jcuw9Pm01duxYi3UKCgoSt2/fLk6ePNnqOVGfbstxZ0t5oiiK7733nsXv7oYNG4p379512DFvS72++OILi3Xy8PAQly1bZvS5Nff+ZGRkiLVr1zZblrnzy6xZs0QvLy+bjitnZ2dx8+bNFre1sMdPfsOHD7epTgDEAQMGiJmZmWbLOnXqlFi3bl2by8v/3bFw4UKbl61fv36B88zu3bttXr5ixYrisWPHTG6HLcemVqsVP/jgA1Eul1tcT6lSpcS9e/eaLcfe71Jbzou2MDxf79q1S5w4caLF7WjTpo2YkpJisUyNRiP+3//9n+js7GzTe+Dp6SmePHnSat1sYe9+vH37tti4cWOrdezatavF7ba3rvbU0zCvtd8RtpS7ePFii9/lSqVS/Pbbb21ar+FxaOn73p7tuH37ttiyZUubjp3169ebLCMrK0t89dVXrX4u9X8BAQFiQkKCxfoTEZFjsKUIET23tmzZgvv372P79u04ePAgTp06hWvXruHhw4fIzc2Fu7s7AgICULNmTbRu3Rr9+vVD2bJlLZY5dOhQVKhQAbNnz8aRI0dw79496Wk1c7777jtERUVhzpw5iImJwcOHD+Hn54fw8HAMGzYMgwcPtnnAaIVCgcWLF2PQoEGYO3cuDh06hMTERPj7+6N27doYOnQoBg0aVKgBqBUKBdq1a4clS5YAcHwrET0/Pz9s3boV69atw9KlS3H06FHExcXB3d0d5cqVQ7du3fDKK6/Y/LRyhQoVcPjwYSxZsgQrV67EyZMnkZCQAG9vb1SqVAkvvvgixowZY7WLgcuXL2Pbtm2IiYnB6dOncePGDamLH09PT5QtWxb16tVD9+7d0bVrV2lA5cetd+/eiIqKwvz587Fp0yacOnUKDx48gCAICAwMRM2aNdGuXTuphZIhpVKJP//8E8uWLcOCBQsQGxuLtLQ0+Pn5oUqVKujVqxdGjx4tdbP1uNWqVQuxsbGYNWsW1q5di2vXriErKwthYWHo3Lkz3nnnHZQvX96h6/Ty8sLevXsxb948LFu2DKdPn0ZmZqa0L/v27YsRI0ZIXZdZ065dO/z777+YMWMGDhw4gFu3biEzM7NIXT84OTlh/vz5GDt2LH777Tfs2bMHd+7cQU5ODgIDA9GgQQP06tULQ4YMMdsS6mmhUCiwdOlS9OnTB/Pnz0dsbCwSEhLg6+uLxo0bY/To0ejevbvVcvr06YPw8HB888032LlzJ+7evQtXV1dUqlQJffv2xdixY41a8VnzxhtvoEOHDvjhhx+wbds23L17F+7u7qhYsSL69OmDcePG2fy5EAQBM2fORL9+/TB79mzs27cP8fHx8PHxQbVq1TBw4EC8/PLLUCgUUhcp/v7+FsuMjIzExYsXsWbNGvz99984cuQI4uLikJGRAXd3d4SFhaFWrVqIjIxEt27dzH5OiuP4BHTjLHXr1g1z5szB4cOH8fDhQ/j4+KB8+fLo0aMHXnnlFQQFBUld4D1uX331FXr16oWffvoJe/bsQXx8PLy9vVGtWjUMGDAAo0aNgpOT02Ot06RJkxAZGYkff/wRBw4cQHx8PDw8PFCmTBl07twZr7zyCipUqICFCxdaLcvNzQ0HDx7EzJkz8ffff+P8+fNITU21OjbFuHHjMGjQICxcuFAawyExMREqlQpeXl6oUKEC6tati7Zt26Jr167w8fFxzMZbMWfOHAwcOBC7d+/Gv//+i4sXLyIuLg65ubnw8PBA+fLl0bx5cwwZMkQaS8Oc8PBwxMbGYtOmTVi3bh0OHTqEe/fuITU1FW5ubggJCUGNGjXQsmVLdO3aFTVr1jRafujQoahVqxZ27tyJw4cP49y5c7h9+zYyMzPh4uKC0NBQNGjQAH369EHv3r0LtPpq3bo1Tp8+je3btyMmJgZnzpzBzZs3kZ6eDicnJwQHB6Nu3bro2bMnBg4cWKTfFoIg4PPPP8fQoUPxyy+/YOfOnbh58yYyMjLg7++P8PBwdO/eHaNGjYKbm1uh1/O4fP311+jcuTN++eUXHDp0CHFxcfD09ETdunVt/v0sk8kwdepUjB07FvPnz8euXbtw9uxZPHz4EFqtFj4+PqhUqRLq16+P9u3bo3PnznB1dX1MW2isdOnSiImJwdq1a7Fq1SocPnwY8fHxkMvlCAkJQYsWLTB48OAn0pKluAwePBj16tXDd999h127duHu3btwcXFB6dKl0b59e4wZMwa1a9cuche0hVW6dGlpTJOVK1fiwIED0rWfl5cXKlWqhGbNmqF3795SN8T5ubi4YPbs2Xj33XexcOFC7N69G5cvX8bDhw8hk8ng6+uLKlWqoGHDhnjhhRfQvn17m39zEhFR0QhiUa+CiIjouZCVlYXSpUsjKSkJvr6+uHPnzhO7MCQiet6dOHFCGgfh7bffxvfff/9kK0RE9AQNHz4cixYtAgDs2rULbdq0ebIVIiIioucaB1onIiIAwLJly5CUlAQAGDlyJAMiRETF6IcffpCm27dv/+QqQkREREREVMIwKEJERNIgpICue5vXX3/9CdeIiOjZdOvWLVy6dMlsuiiK+Oqrr6RukSpUqIAuXbo8ptoRERERERHR093xNBERFZs9e/YgKysLd+/exezZs3Ht2jUAwKhRoxw+bgMRUUlx5swZdOnSBY0aNULbtm1RtWpVeHt7IyMjA+fPn8cff/yBCxcuAND1dz9//vwC4xAQERERERFR8WFQhIiohBo2bBhu3LhhNK9q1ar48ssvn1CNiIieD6Io4siRIzhy5IjZPD4+Pli8eDH7zSciIiIiInrMGBQhIirhnJycUL58efTo0QOTJk2Ct7f3k64SEdEzq1WrVvj999+xbds2nDp1CgkJCXj48CFEUYSfnx9q1aqFjh07YtSoUTzfEhERERERPQGCKIrik64EERERERERERERERFRcWMHxkREREREREREREREVCIwKEJERERERERERERERCUCgyJERERERERERERERFQiMChCREREREREREREREQlAoMiRERERERERERERERUIjAoQkREREREREREREREJQKDIkREREREREREREREVCIwKEJERERERERERERERCUCgyJERERERERERERERFQiMChCREREREREREREREQlAoMiRERERERERERERERUIjAoQkREREREREREREREJQKDIkREREREREREREREVCIwKEJERERERERERERERCUCgyJERERERERERERERFQiMChCREREREREREREREQlAoMiRERERERERERERERUIjAoQkREREREREREREREJQKDIkREREREREREREREVCIwKEJERERERERERERERCUCgyJERERERERERERERFQiMChCREREREREREREREQlAoMiRERERERERERERERUIjAoQkREREREREREREREJQKDIkREREREREREREREVCIwKEJERERERERERERERCUCgyJERERERERERERERFQiMChCREREREREREREREQlAoMiRERERERERERERERUIjAoQkRET4Xy5ctDEAQsXLjQYWUKggBBELB7926HlUlERERERM8GXmMQEZEpDIoQET3jpkyZIv0wN/xzdnZGaGgoOnfujIULF0Kj0TyR+v3www+YMmUKjh8//kTW/zxy1D69f/8+Vq9ejQ8//BCdO3dGcHCwdPw48sKRiIiIiJ4tvMYoeRy1T69cuYLvv/8ePXv2RIUKFeDi4gI3NzdUqlQJgwcPxp49exxTYSKiIlA86QoQEZHjBAcHS9NpaWm4d+8e7t27hy1btmDu3LnYvHkzvL29H2udfvjhB9y4cQPly5dHvXr1zOarVKkSXFxcHFq/atWqAQDc3NwcVubTwNZ9as2cOXMwdepUx1WMiIiIiJ47vMYwxmsM8w4cOICWLVsazfPw8IBarcbVq1dx9epVLF26FGPGjMHs2bMhk/FZbSJ6MhgUISJ6jty/f1+aFkURV69exYcffohVq1bh0KFDePXVV7F8+fInWEPzduzY4fAyz58/7/AynyeCICAsLAz169dH/fr1UadOHfTp0+dJV4uIiIiIniK8xjDGawzzVCoVZDIZunXrhkGDBqFNmzYICgqCVqvF2bNn8cEHH+Dvv//G3LlzERISwge0iOiJYUiWiOg5JQgCKlWqhBUrVqBFixYAgJUrVyIuLu4J14yeFh999BFu3bqFDRs2YOrUqejZs+eTrhIRERERPcV4jUGWVK5cGefPn8f69evRr18/BAUFAQBkMhnCw8OxYcMGdOzYEQDw3XffITs7+0lWl4hKMAZFiIiec4IgYOjQoQB0T3b9+++/AIDMzEysWLECw4YNQ7169RAUFAQnJyeEhISgS5cuWL16tdkyd+/eLfUrDADHjx/HkCFDUKZMGTg5OaFevXpSP8Q3btwAAIwYMaJAn8TXr1+XyrRlEMTY2FiMHj0a1apVg4eHBzw9PVG9enUMGDAAa9euhVarLbDtpgZBvH79ulEdrl69ilGjRqFs2bJwcXFB6dKlMWzYMFy+fNlsXU6fPo2pU6eiffv2qFSpEtzc3ODp6YnatWtjwoQJuH37ttll27RpA0EQMGXKFIiiiHnz5qFJkybw8vKCp6cnmjVrhiVLlhRYzt59ao1CwQajRERERGQ/XmPwGsOUsLAwVKlSxWy6IAh4+eWXAQDp6elsdUNETwzvhhARlQBhYWHSdEpKCgBg1apVGDFiBADdj1MvLy84OTkhLi4OmzdvxubNmzFgwAAsXbpUujAxZd26dejfvz9yc3Ph6ekp3Wj38PBAcHAwEhISoNVq4eXlBVdXV6Nl5XK5TfXXarWYNGkSpk+fLs1zdXWFi4sLLl26hAsXLmDFihVISkqCj4+PTWXqHT16FKNGjUJqairc3d0hl8tx9+5d/P7771i9ejU2bNiADh06FFiuW7du0oWDs7Mz3N3dkZycjNOnT+P06dNYsGABNm7ciGbNmpldt0ajQa9evbB+/XooFAq4ubkhLS0NMTExiImJwaVLl4yalDtynxIRERERFQWvMczjNYZ5Li4u0rRarXZYuURE9mBLESKiEuDatWvStJ+fHwDA19cX77zzDvbu3Yu0tDQkJycjPT0dCQkJ+O677+Dp6Ynly5dj1qxZFsseOnQo2rZtizNnziA1NRWZmZlYtWoV3n33Xdy/fx9lypQBAPz444+4f/++0Z8+zZrJkydLFysDBw7E6dOnkZmZiYcPHyI1NRWbN29Gv379CjVQ35gxY1CmTBkcOHAA6enpyMjIwL59+1C9enVkZWWhb9++uHv3boHlIiMjMX/+fFy7dg1ZWVlITExEdnY2Dh48iKioKCQlJaFv374Wm4TPmjULO3fuxPz585GamoqUlBTcunUL3bt3BwBMmzYNly5dkvI7cp8SERERERUFrzHM4zWGefpxXpydnaVB64mIHjuRiIieaZMnTxYBiOZO6SqVSqxdu7YIQJTL5WJiYqJN5a5atUoEIFauXLlA2q5du6R1RkREiCqVymw55cqVEwGICxYssLg+c/kuXbokyuVyEYD41ltv2VR3PX0dd+3aZTT/2rVrUpqvr6947969AsveuXNH9Pb2FgGIr7/+ul3rVavV0j5fsmRJgfTWrVtL69+2bVuB9OzsbDE0NFQEIE6bNq1Auq371F4qlUqql6PLJiIiIqJnB68xzOM1RuFdunRJdHV1FQGII0eOLLb1EBFZw5YiRETPqdTUVBw4cAAdO3bEqVOnAADDhw+XnuKyplu3bgCAy5cv4/79+2bzvf/++8U6NsWiRYug0Wjg4+ODzz//3OHlv/rqqwgJCSkwPzQ0FGPGjAEArFixwq4y5XI5OnfuDADYt2+f2XzNmzc32Wze2dlZGoDw5MmTdq2biIiIiKi48BrDNrzGKCgjIwN9+vRBVlYWgoOD8dlnnz32OhAR6XFMESKi54ilfnk7duyIGTNmGM1LSEjA7Nmz8c8//+DChQtISUkx2a/r7du3Tf6oB4CWLVsWrdJW7N+/HwDQvn17uLm5Obx8UxcMeu3atcP06dPx4MEDXL9+HeXLlzdK37x5M37//XccPXoU9+/fR0ZGRoEyLA2G2KRJE7NpoaGhAICHDx9a2QIiIiIiouLDawz78RrDWG5uLvr27YsTJ07AyckJy5cvN/veExE9DgyKEBE9R4KDg6VpJycn+Pv7o06dOujbty+6du1qdEETExODLl26ICkpSZrn7u4OX19fqd/cuLg4ADD5Q1wvKCjI0ZthRP8EWf6LBUcpXbq02TTDwSPj4uKkOmi1WgwbNgxLliyR0uVyOXx9feHk5AQAUt/Blvadl5eX2TT9k3Eqlcqm7SAiIiIiKg68xrAfrzEeyc3NRZ8+fbB582YolUqsWrUKbdu2fWzrJyIyhUERIqLniKUm6IbUajUGDBiApKQk1K5dG59//jlatWoFb29vKY9Go5F+NIuiaLYsuVxetEpbYenJtCdl/vz5WLJkCWQyGd5//30MGzYMlStXNtoXn3zyCaZNm2Zx3xERERERPe14jfF4PI/XGDk5OYiOjsbGjRuhVCqxcuVK9OzZ80lXi4iIQREiopLo0KFDuH79OmQyGTZu3IgyZcoUyHPv3r0nULOCQkJCcO7cOVy/fr1Yyr9z5w6qVatmNk3P8Am55cuXAwBefvlls30QPy37j4iIiIjoceA1xiO8xgCys7Px4osv4p9//pFaiLz44otPulpERAAADrRORFQC3bp1CwAQGBho8mIFALZt2+aQdembyRf2aaYWLVoAAHbs2IHMzEyH1MnQzp07rab5+/sbNa3X778GDRqYXE6r1Vost6iKuk+JiIiIiByN1xiPlPRrjKysLHTv3h3//PMPnJycsGbNGgZEiOipwqAIEVEJ5OPjAwCIj4832Rw+JSUF06ZNc8i69M3lDfsVtsfw4cMhl8uRnJyMDz/80CF1MjRnzhzEx8cXmH///n3MnTsXANC/f3+jNP3+O3HihMkyf/75Z1y7ds2xFTVQ1H1KRERERORovMZ4pCRfY2RmZqJbt27Yvn07nJ2dsW7dOvTo0cMRVSQichgGRYiISqAWLVrAw8MDoiiib9++OHfuHADdU0F79+5F69atkZKS4pB11a5dGwCwZs0aJCYm2r18pUqV8P777wMAfvzxRwwePBhnz56V0jMyMrBhwwb06NEDqampdpev0WjQoUMHxMTESPMOHDiA9u3bIzk5GV5eXpg0aZLRMp06dQIAzJs3D7Nnz0ZOTg4AIDExEVOmTMFbb70Ff39/u+tiq6LuUz2tVosHDx5If4ZlpaenG6UVxxN0RERERPT84DXGIyX1GkMfENm5cydcXFywfv16dOnSxZHVJCJyCAZFiIhKIG9vb3z33XcAgP3796NmzZrw9PSEu7s7WrdujStXrmD16tUOWdeYMWMgk8lw6NAhBAcHo1SpUihfvjzKly+P27dv21TGp59+irfffhsAsHTpUtSqVQvu7u7w9/eHl5cXevbsib/++gtardbu+s2dOxc3btxAs2bN4OHhAQ8PD7Rs2RJnz56Fi4sLVq1ahbCwMKNlJkyYgJo1a0Kj0WDcuHFwc3ODn58fAgMDMXXqVHTr1g1jx461uy62csQ+BYCbN28iMDBQ+gsJCZHS3njjDaO0r7/+ujg2hYiIiIieE7zGeKSkXmOsWbMGu3btAqALhg0bNgwhISFm/1auXFls20NEZAmDIkREJdTo0aOxZcsWtG/fHl5eXlCpVAgODsbo0aMRGxuLtm3bOmQ9LVu2xObNm9GpUyf4+voiISEBN27cwI0bN6BWq20qQyaT4fvvv0dMTAyGDh2KChUqQKvVQq1Wo0qVKhg4cCDWrVsHLy8vu+vXqFEjxMbGYuTIkfDz84NKpUJISAiGDBmCEydOoGPHjgWW8fb2xsGDBzFhwgRUrFgRcrkcgiCgRYsWmDdvHv7880/I5XK762IrR+xTIiIiIiJH4zWGTkm9xjAMIOXk5CAuLs7iX1ZWVnFtDhGRRYLIUVqJiKiEuX79OipUqAAAuHbtmtEAh0RERERERPbiNQYR0bODLUWIiIiIiIiIiIiIiKhEYFCEiIiIiIiIiIiIiIhKBAZFiIiIiIiIiIiIiIioRHjmgiLHjh3D119/jb59+6J8+fIQBAGCIOD69esWl7ty5QqGDBmCUqVKwcXFBZUrV8akSZOQnp5udhlRFDFnzhxERETA3d0dfn5+iIqKwo4dOxy8VUREREREREREREREVNyeuYHWX3zxRaxfv77AfEuDWB07dgxt2rRBWloaGjRogMqVK+Pw4cO4ceMGwsPDsX//fnh7exstI4oihgwZgqVLl8LT0xMvvPAC0tLSsHPnTmi1WsybNw8vv/xycWwiEREREREREREREREVg2eupUizZs3w8ccf488//8Tt27cRHBxsMb9Go8GAAQOQlpaGL774Av/99x9WrlyJixcvolu3bjh9+jTee++9AsstWbIES5cuRYUKFXD+/HmsXbsWW7duxc6dO6FQKDBu3DjcuHGjuDaTiIiIiIiIiIiIiIgc7JlrKZJfSEgI4uLizLYU+eOPPxAdHY3w8HCcPHkSgiBIaffu3UPZsmUBAPfv34e/v7+UVqdOHZw6dQqrV69Gnz59jMp87bXX8PPPP+Ptt9/G999/XzwbRkREREREREREREREDvXMtRSx119//QUA6Nu3r1FABABKlSqFVq1aQa1WY+PGjdL869ev49SpU3BxcUGPHj0KlPnSSy8BgMluvIiIiIiIiIiIiIiI6On03AdFjh8/DgCIiIgwmd6gQQMAwIkTJ6R5+unw8HA4OTmZXebatWtITU11ZHWJiIiIiIiIiIiIiKiYKJ50BYqbftyPsLAwk+n6+Ybjg1hbxsPDA97e3khJScHNmzcRHh5usQ7NmjUzOf/q1avo0aMH5s2bZ3kjiIiIiIiIDPAag4iIiIiocJ77oEh6ejoAwN3d3WS6h4cHACAtLc3mZfTLpaSkGC1nr8zMTJw4cUJaHxERERHR80r/u5uKF68xiIiIiKikKOw1xnMfFHkaHDp0yOR8/dNdvEAkIiIiIiJ78BqDiIiIiKhwnvsxRfQXAxkZGSbT9U9QeXp62ryMueWIiIiIiIiIiIiIiOjp9dwHRcqVKwcAuH37tsl0/Xx9PluWSU9PR0pKCgCgbNmyDqsrEREREREREREREREVn+c+KFKvXj0AwH///Wcy/dixYwCAunXrSvP006dPn0Zubq7ZZSpUqAAvLy9HVpeIiIiIiIiIiIiIiIrJcx8U6d69OwBg9erVEEXRKO3evXvYt28fFAoFunTpIs0vX748ateujezsbGzYsKFAmStXrgQA9OzZsxhrTkREREREREREREREjvTcB0V69OiBqlWr4vTp0/jqq6+k+bm5uXjllVegVqsxcuRIBAQEGC03ceJEAMB7772Hu3fvSvP37t2LefPmwcnJCW+//fZj2QYiIiIiIiIiIiIiIio6QczffOIpt3HjRnz66afS62PHjkGlUqFevXpwdnYGAHTt2hWffPKJlOe///5DmzZtkJ6ejoiICFSuXBkxMTG4ceMGwsPDsX//fnh7exutRxRFDB48GMuWLYOXlxc6dOiA9PR07NixA1qtFnPnzsWoUaOKtC3NmjUDABw6dKhI5RAREREREQG8xiAiIiIiskbxpCtgr4SEBBw+fLjA/OPHj0vT1atXN0qLiIhAbGwspkyZgu3bt+PUqVMICwvDe++9h08++QQeHh4FyhMEAUuWLEHLli0xb948bN68GU5OTmjbti0mTZqE9u3bO3zbiIiIiIiIiIiIiIio+DxzLUWeJ3yKi4iIiIiIHInXGERERERElj33Y4oQEREREREREREREREBDIoQEREREREREREREVEJwaAIERERERERERERERGVCAyKEBERERERERERERFRicCgCBERERERERERERERlQgMihARERERERERERERUYnAoAgREREREREREREREZUIDIoQEREREREREREREVGJwKAIERERERERERERERGVCAyKEBERERERERERERFRicCgCBERERERERERERERlQgMihARERERERERERERUYnAoAgREREREREREREREZUIDIoQEREREREREREREVGJwKAIERERERERERERERGVCAyKEBERERERERERERFRicCgCBERERERERERERERlQgMihARERERERERERERUYnAoAgREREREREREREREZUIDIoQEREREREREREREVGJwKAIERERERERERERERGVCAyKEBERERERERERERFRicCgCBERERERERERERERlQgMihARERERERERERERUYnAoAgREREREREREREREZUIDIoQEREREREREREREVGJwKAIERERERERERERERGVCAyKEBERERERERERERFRicCgCBERERERERERERERlQgMihARERERERERERERUYnAoAgREREREREREREREZUIDIoQEREREREREREREVGJwKAIERERERERERERERGVCAyKEBERERERERERERFRicCgCBERERERERERERERlQgMihARERERERERERERUYnAoAgREREREREREREREZUIDIoQEREREREREREREVGJwKAIERERERERERERERGVCAyKEBERERERERERERFRicCgCBERERERERERERERlQgMihARERERERERERERUYnAoAgREREREREREREREZUIDIoQEREREREREREREVGJwKAIERERERERERERERGVCApHFJKZmYndu3dj9+7dOH78OOLj45GcnAwfHx8EBQWhfv36aNOmDdq0aQNXV1dHrJKIiIiIiIiIiIiIiMguRQqKnD17Fj///DOWLFmCtLQ0iKJolH7z5k0AwPbt2/HNN9/A09MTQ4YMwdixY1GzZs2irJqIiIiIiIiIiIiIiMguhQqK3L17Fx999BEWL14MrVaL0NBQdO3aFU2bNkWNGjXg7+8PLy8vpKSk4OHDhzh79ixiYmKwZ88ezJo1C7Nnz8bQoUMxbdo0hIaGOnqbiJ4ruWkpAAAnT+8nXBMiIiIiIiIiIiKiZ5sg5m/eYQN3d3eoVCoMHDgQw4YNQ5s2bSAIgtXlRFHEjh078Pvvv2PFihVwdnZGWlpaoSr+PGjWrBkA4NChQ0+4JvS0yk1Lwf6P3wAAtJw2k4ERIiIiIrKI1xhERERERJYVaqD1l156CRcuXMDChQvRtm1bmwIiACAIAjp06IDff/8dFy5cQN++fQuzeqISQR8QSbl6ASlXL2D/x29IrUaIiIiIiIiIiIiIyH6FailCjsGnuMgcw4CIIe+K1dhihIiIiOyi0Wqx9OA2AMDg5i9AJivUc1H0jOA1BhERERGRZbwiInrKmAuIAGCLESIiIrLb5hMxmLx2PiavnY9NJw4/6eoQERERERE9UcUaFMnOzsaZM2dw9uxZ5OTkFOeqiJ4LlgIiegyMEBERka00Wi1m/LNWej3jnzXQarVPsEZERERERERPVrEERVQqFd577z34+vqiTp06qF27Nvz8/DB58mRehBGZYUtARI+BESIiIrLF5hMxuBR3W3p9Ke42W4sQEREREVGJpiiOQt955x3MmTMHAwcORJMmTZCZmYnly5dj2rRp0Gg0mDZtWnGsluiZpQuIvI6UqxdtXkYXGHkdLaf9xDFGiIiIngNarRYqjRq5GjVUajVUGt1fbt60WqMxm6YymJ+rVkOt1SBHlYv5ezcXWM+Mf9agS90mHFuEiIiIiIhKpGIZaN3T0xMTJ07E//3f/0nzVCoV6tSpg9TUVNy5c8fRq3wmcRDEkkWrUSM78QEyE+4hKyEOmQn3kZUQh/S7N5F49iS0qtxClStTOsGvRh14hJaBW1AI3IJKwS04FO5BpeDs4weBNzyIiKiEEkURmrxAg1EAQR88yAs0mEozGXDQakynG5Snyh/AsBrE0EivNY+xRfXMoW+jW/1mj2199PjwGoOIiIiIyLJCtRTp1KkTfv75Z1SsWLFAWnZ2NjIyMlC/fn2j+UqlEjVq1MDmzQWfViN6Hqgy05EZfx9ZCfeRmRAn/c+M1wVBsh4mAMVws0OrysWDk//iwcl/C6TJlE6PAiWGf8G6/y6+/gyaUImj73aOLayICkcUxUc38k0EA6SWChrDNI1RIMBUsECl0VgsTx9oMNs6wlR5GjWK4fmf58KUNb/jhfBGcFYWS8NxIiIiIiKip1ahroJOnDiB8PBwvP/++/jggw/g5OQkpbm4uKBSpUqYNWsW2rVrB3d3dwDAsWPHsGPHDtSqVcsxNSd6jESNBlkPH+QFOnQtPDLj86Yf6AIgqox0h69XC+CgtycAoHlKmt2DAGlVuUi/cxPpd26aTJcplHANCoG7PlAS+Chg4hZcCi6+AQya0HNFP3YPALScNpOBEXpqaLRaqC0EAx7N0xRMy9dtkn6ebV0taSwEFwoGINR5daBnX2LmQwyYPhsr338NSjm/64mIiIiIqOQoVFDk4sWL+OijjzBt2jQsXboUM2fORMeOHaX0r776Cv369UNoaChq1KiBjIwMnDt3DoIg4LPPPnNY5YkcRZWZYdCllWFLD10AJOtBPESt428CKd094BoYAteAYDj7+CH+2CFkP3wgpZ/wcMMfgf4AAA+NBvXSM6U0F/8ghDRqgZzkh8iMv4fMuHtQZaTZtX6tWoWMu7eQcfeWyXSZQgnXwGC4BYfCLagU3INKwTWv5Ym7Pmgilxdiy4kev9y0FOz9+HVsTrwHANB+/DoiOSbPc8uWbpMKBBosBA9yDQIN1suz0tWSie6bHme3SeQYSrkcSrkCSoUCCpkCTgqF7rVcDqVCASe5QkpXyo3T9K+dDNPyLaOQyR+lW1hGKXtUpj5NnZGOXl++iwS5YHEbbt7Zg1/XN8DY3i0e014jIiIiIiJ68oo0pkhsbCzGjRuHI0eOoHfv3vjhhx9QunRpALrWJNOnT5eCIeHh4Zg4cSJbihhgf7+Ph6jRIDs50TjQEW/cxZUqPdXxK5bJ4OoXqAsqBJWCW2AwXAND4Jb32jUgGEp3D6NF9E+xp1y9AC2A6WVCEeesa4kVnJOLibfuQgbAu2I1k0+5qzLSdQGSvCBJhsF0Zvw9h2+noFDALTDEuFuuwJC81iahcPVj0ISeDvrP1u74W1gcEgQAGHo/Hq2DyrDFiB1s7TbpUfdF1rtNksZ0sKU8qy0bjNdDzx4nhVIXGMi7ua+Qy00GF4yDBbo8CjPBBX3wwkkqU2G0Dl0+0+sxXZ6uTADQaEXkqjVQabRQqbV5/wu+zrWSbvg6V63Jm9YY5Hn0Ojff6/zpspwM1Lj7BzYEuNi0z3sk5uDzGQvg7uNbnG8tPUa8xiAiIiIisswhA63PmzcPH374IXJycjB58mS8/fbbkPNGqFVP+oLleelXX52d9ahLKyngYfA6MR6i2vE3xxSubrrgRmAw3AJDDP6HwC0wBC7+AZDJ7W+MZermrV5Rb+Lqxz0xDJRkxt1FZrwugKJKc3DQRC5/FDQJLlVgbBNX/8DnKmjyvHymnjf6z1TS1QsmA42+ZoKMj4ut3Sbp09VajdVuk3RBAY00poO54MKjcRqsd5ukXzc9W+QymfkWBvqWDgY3+/MHC5zyWizkb4lgahmjlg2G6flbSggyXb1kMiiEvP8yGWQQIECEVqPRBQpU6kf/VWpdsEClhlqtm69W6YJparVG96fRPErL+9No1FCrtdCo1VBrtNBo1NBotFDnvdaqNdAYzNdodfO0Wl3rHa1GC41GA61WC61GA61Gq5vW6uYJoggBIgRAmpZBNJj/KF0manX5DNJl+nTDciBCln/5fK9lIkymy0UNglWJmBEWJJ3rrAnOycWHWmd0mv4Lv7+eE0/6GoOIiIiI6GnnkKAIADx8+BATJ07EwoULER4ejp9//hktWrApviVP8oLlWelXX9Rqdd1DSeN43DMexDzB8TfyAeS18giQWnY8+v+oxYfS3QOCYLlbCluJoois3Bxk5GQjIycbiYlxeGXO50jMV7yfCMx4eSJ8/QLg5uQCNydnuDo7w83JBXIHjP3xKGhy3yhgog+i6G/6O4ogl8M1IFgKmOi65yoljXHi4h9YqMDSk/CsfKZKGn2XWQ+uXcIxT3esDgowSo96mIwKWdlwLhWGqgNHQ3RyMtttkrmulsy1UrB1UGgtB4F+5hjeyFfK5FAI+mlBNy3IoDCYVgoCFIL+dd40BMgFQAkB8rzXuvmAQhAgR14aAIUIyCFCjrx0iJCLAuTQQiHq03XzZVoRclGEHCIUWt1NdlGrAbSidDNf1Obd2Nfob+6L0jxRo4EoaiHm5YdWm/c6708Ugbx0iFqD6UevBTHvNXT/BVGEIGohgMf643DC3Q1LQgLtWmbw/QREBrPl3POCQREiIiIiIsscFhTRO3ToEMaOHYtTp05h6NChmD59OgICAqwvWAI9qQsWwy6aAPNdMT0O6uxsaaDyTMMWHg/yWnw8iCuWVh5yF1ddCwZ9C498LT5c/YMgU5i/Ga/WaJCRkyUFMXR/j15n5mQjPScLmWbS8+fNyM1GUT+KzkplXqDkUbDE3ckFrk7OcMsLnLg5uRhMG+Yxnu/mnPffyQXOSqUU/NGNvXJfCpRkxBl3z5WbmlykbchPkMnhGhBkMPh7aF4rk5C8LsiCnoqgSf6xKjr7l+JYFXbKVauRrcpBZk4OMnOzkZWbg8xcg+mcnLx5utdZublG+fRpmbk5yMrJWzYnCxkZ6ch1TOySipEgipADkIsiFGLeTX0RBtNm0mGQbpgf+ZfPm87LrxBhUG7BdUnp0K8PRvXgIUXPoyf5e5Aci0ERIiIiIiLLijymyPHjx/Hw4UOUK1cObdq0QUBAALRaLWbMmIEpU6ZALpfj888/xyuvvOLIej8XHucFy6g3PoBao4ZSo0LzhFPwUqXjoLcnAKB5ShpSlR44GFgbKrkSCrkCv878osjrFEUROckP87qxMm7hkZXX4sPRN9EBAIIAF98AoyCHzD8Q8PGF1ssHWncP5MjlyMzNLhDU0AUz8gIVJoIe+vRctcrx9X5KyQQBbnnBFXfnvCCLQeDE1clFmu8ik0GZq4I8JwuyjAzIMtIgpKQAyQ+hTXoAITUFzloRTqIWTlrdzcWikIImhmOaSEGTUF3QxEJwqyjyf6auOWuNxqoonyNz+GfqSdK3ZspS5SIrNzsveJGjm5aCEjnIzDEIVKgeBSgK5NPPywt2sFsmx5PZEBBQGAQbrAUEFAZBgQJ5RTEvQGEpb750g2BE0du5ET2FBCHvTwYIgu4BA5kcgiBAkOXNk8kgCDIIct1/mUxmnCaTAYJMWkZ6LRMgCDIg779+vqjVIPXaZaizM+2qqlfFqmjFYP5zg0ERIiIiIiLLCnW3MCUlBYMHD8amTZuMnm738PDAtGnT8Oabb+Ltt99G//79MX78eIwdOxYLFizA7NmzUb9+fYdVnmyn1qgh5GajWeIZ+KgzEOvhhj8C/QEAHhoN6qWno2HCWRz0qwa1k20Dc2pyc5D1IM64hYfhmB4P4qFV5dpdVw2AXJkM2TIBuULef5kMOTIBOYIMuU5OED29oXF3h9rVDSpnZ6iUSuTIZMgRgCyNFpmqHF0QI/kOMuKusHuaItCKItJzspCek4WEtCIU5Oup+zOgFAQ4Q4CTKMJJrYZSrYaTKMJZqwua6IMnTqJWF0wxmOes1eqWS3kI5+QHcDp/UsqnFHVdxkAmg6t/kBQwcZcGg9f9dw0ILnTQxPAz5aXOwNaQUCntH18fTLx1F83iT2Kffy2bP1NFpdKo81pRGLSeyDEORuiDE/oWFVkqg1YWOQXzGZZFBVsqWAwI5LVkkFkJCBgGGGQWAhFGrSbytZLIH+iQAQw0PGO0wKMRIgTj0ShEAKIgQGsiXWv0GlK61nD5AvlhXH6+9WkF5I12AaP0guvXpSPvZr3+hr7+Jr9MkEEmN0yTQybXTctkMsjkcshlMsgUef+l1zLI5QrI5TLIZXLIFTLI5XLI5XIo5HLIFXnTCjnkchkUCt2A7AqFXDetkEGh1M1TKnSDqMsVcl0d5HlBiXwBBSkQIch02yMIEGRyCDLBKAhhGJSQ5ucFOqDfdmn6ybQpyk1LwdJhveGjSrcpf4LSG10ZECEiIiIiohKkUHcDx48fj40bN2L48OEYPXo0/Pz8cO7cOUyePBnjx49H3bp10bp1a4SEhGD58uUYNWoUXnvtNTRu3BivvvoqZs6c6ejtICuUGhWaJZ6BrzoDWgBbfX2ktH98fVAnPRNBqhQ0fXgJ/wZWhyiKyE1NNgh4GLfwyHpwHznJSRABqARBCljkyARdcEImQ46TArkuTlJQQx/kyJEZ5BOEAq/Vto6NkZuh+3vKuTk5w93ZFe7OuhYVbs4u8HB2hVvea92fLn3Nuk24Ls+ARmalJYrogkB1EN4aEIWM3Gxk5WRLXQ3pn+B/dJM7GxkGT+Vn5mZDo9U+no23QiWKUOn7mFfIdX8OIIiiLmCiD6KkxsE5+R6czotw0gdTtCKcRRHuLq7wcPeEp6c3PL394O3nDx//IPgEhsA3sBQ83D3yuhLTtY5R5nXXZfiZivVwMxrQNs7ZCSc93FAvPQOtEs/gUFAdALrWFtmqXCnIYNgdlNmWF7nZ+YIXOdL7mj/YodKUnNYWclH3XupaHhm+r1rj914r4qFSgZMe7hbL6/LgIWpmZj0KdJgITsjw9HSbpL8xDSHfjW1BgAYC1NJNcUg3s/WvH90sB7SQWbxBbu0Guqkb8lrB4GZ6/hvqgmA+zShdJgUCTN+Mz6u/IDNat/nyTNQr/7aYqZtxunGQwlzdzG1rwUDFo21FvhvoggA4KeRQymVQKmS6AcoVMpOvnfKny2VQ5i3rpDSzrDRtkDff6/zpSoVMVyeD+QqZ8MRu/pNpTp7eOBhYG83iT8JXbfl30n2lL/4NrI4xDIgQEREREVEJUqjus/z8/FCzZk3s37/faP7ly5dRtWpVvP322/juu++M0lQqFb7++mt88cUXSE+37cm1593jatqe/4nBWA83qZsfvTYPk1E6V4UcmYAsmRwqhRy5oohsmQy5BQIYMqMgiPgc3QwRIINCroRCpoRcpoRC0P2Xy5SQCwrpv0x49F8mKHR/UELImxYEBWSiArqhcQUAIrRaEbqxaEWIoq4FhlYUIYqP5t9+eA45ytM21VWpqovx3brC39MV3m5O8HF3ho+7C3zcddPOStMxT1EUkatRIysnRxdQMQiWGAdUdDfrM/JuykvzpdYH2dLg8CWpNYFCkMFVoYSQlQkXrRZKrRbxTkqo8gXzlFot/FRqqGQCcmRyiK6uyFLlFnnsmGeFIOpa7Bi3+DFo4aN9NM9UIMNUoEPXWkgLpWjQ9ZogQOHiCrmzi+5/3rQgVyDl+iWoMjMwvUyoUdDKlOCcXEy8dRc5br64XrktsuVO0GhFqLSAWou8/9q8aVH6r9Lo8qg0ovSkv7bADXWYDBSYvkFu/ea9qRvoZDu5TMgXbLAcBCgQFDCRnj9QYEtZ5gIQ+rLktj4gQGTC8HETIeRmo1Ve8N6UeKU3DvpVg+DkgoU/T3/MNaTixO6ziIiIiIgsK1RLEZVKBQ8PjwLz9fNUqoJPuSuVSnz00UcYPHhwYVZJhaQfVF0fEMnfSkRvt1/Bec8EUQ5ADgGKvGlFvmndf4h5eYym8/Lk5dWVo7vVqs37s3/kEE3en/3BAREiVIrrNudXy6/gh7+OQTDz7LpcEOEsE+EsF+EkA5xlIsKrlkNYkF9eAEX3J2hysWnzVimfzERxAgB3AO6QA3BDqeByeHlIvwL5/vpnJ/49cRpqUQOVqNH912qNXuefV7Zsabh6uBgFV9Kzs3AnLs4orxZPRzBBLWqRpsoBFHKkWhgVRSWTGd+IfwoDRvL8AYl8wQnnfEEKpWFwIn/wIn9efRdmyLvRL1fo/mQKiNK0UvdfoUCValXh5esLhasr5M6uULi4Ii1Hhd2HY5EpKJEkKJAtKJEtKJElKpElKJEFJTK1CuSKMuRqgRytAK2ghChTIjUjF86+9dFCudlqQATQte7Z7R2C/R6dkZVsR3dn+n6qimfommfKoxYKuv/Z2VkQRBEyAZAJef8ByPO9NkxvWLcWXJyURi0gEh88xLkLF/Py6fLKpeXyljV4LRdEhFevjA6tm0EplxsEH2T4fflaxMU/sL4xIgC17m/syIHw9/M1Sr5x6w4Wr/wTGgDZNu6fKhXL4aXe3QrMX/nH37h09YaNpegMeelFlCtT2mhe4sMkzJ6/zK5ySgUHmjyXb9y6C7Enz9pV1otdXkB4zapG83JycjF95jy7yvF0d8NbY0cUmL/34BHsPXjUrrI6tG6Opo0Kdtv69Y9zkWvit6olH7/7WoF5x06cwaZtu+0qp0nDenihTYsC83/+bQkeJqXYVdZbrwyDp6fx7/GLV65h1bpN0Gg0EGVK7POvZTIwkqRwR4xfNWhkSp6+iIiIiIioxCnUdVCHDh2wYcMGfPnllxg5ciR8fHxw/vx5vPvuuxAEAe3btze7bLly5QpdWbKPPiCScvWCNO9Evm5+HitRgBSIgBwQFfmmTQQ38tIeBTcMpiE3GxB4FmmFOIhCRt5+sk4UMqAV4iAXQ0yma0QBmRoBmQY9Kt08fgfAHRO53aQpuaiBHGooRA0U0ECRNy03mA5N1qLR9QR4uznDx8MZni5OkMkExMc/wJXL163WXZn35wpgYLfWaN4kwig9Kzsbo9/80GieFiK0EKERRGighVYQoYEo/Xd2c0Z0r87SGBr61i4Xrl7D9Tt3oRFEaKVStHnliFAJeKqfujcVcDDdsuLRuCtKky0zjPMWGOheEKCCDGpBBo0gh1qQG0zLoBEUUMsevVYLcoRWKod27SLzghgukLu4QuHsil+Xr8OlG3egluUtC91yeUcW1Pppg3l1vepAKyqR8iAHKRm5SMnMQXKGBjmqcDv3mBb6oGSm3Bmb/f0A2Dbg8CZ/PyjUzk/dWUUQdW1KZBDh7uoMdzcXKBVyoy6Trly5JuUxbFMigwhBNOoECjKIUCrkeOnFzgW6RTp95hwOHDqKfO1YIBNNlJv392KX9ujXs1OBLpRenzgFySmpdm3r+72GwNXFOCh18PB/OLf3ghSstkWNMgEI8/csMP/GzTu4fvO2XXUydfM8IzML5y9esascVxdnk/Nv371vd1kZmVkF5uWqVHaXk51tOlh7Py7B7rJSWzUtME8rau0ux8fby+T8hAcP7S6rQZ1aJudfvHwN2TlFD1Qnp6TaXacK5cJMzr967RbuxyfYVZbaRJeJ6emZRnVSmQiMJCncsc+/FlQypV3rIyIiIiIiel4UKigyc+ZMXL58GR9++CE++ugjo7SxY8fixRdfdETdyMHMtRIxR6kVoev+SQkRCmjhBEAOwTCAkb/lhVHQwrgVBiBzfBAj72l0hUIOmUxXukwQIMgEaDUa5ObqBnrXrVU0WLvpaTdXF/h6e+nKEABBECATdDeI8pdlbbp6lYpwdXE2Kic9IxMXLz26WWG8jDNUmuZIlZm+IfS46G9e51p4q64lAQemb5ReywQBXm5OEDS5yJVXkoIp8nyBFQU0UIjqvACLBjKbb3Hqjh4ZBChEQHesGaf7yLwwtGXHAsut3bAF665uNVuuKIVJRIjQQKnNQdM6VREa4ImUxAdISX6A1JQkpKelID0rE7l5Y+PkygTEKZW44Wq5VUGFzGyEqFQmu4JS5gt0OOdrsWHY2gIABJkccheXR91FubniVlwistRaXaBCJocmL2ChyXudJZcjzXCeIMf4t16Fb2AAFM667qYULi44fvYSvv95vsVt0UKA7t3TBTRS/KvCx6MGUjJzkJKcg5TMXKRkpuNYWjBSnX11gZW8/NYCT1tP3bOYXhhaIQ5aIQsQBcitHGsayKAVsiAT7sBF6w9BFFG6VCB8vTyNuj2CVoNjx0/qAgJiwQCBqWCEr7cHhg/oZRDI0JW1Y9c+HDr8X74ytEZlAMbjmIzs2RftIpsVqP/gMe/YtW9cBGeM7FAw4CRPuokzon2BDI4pQfR0MwyMAGBAhIiIiIiISrxCBUXCwsJw8uRJrF27FsePH0dSUhLKli2LLl26oHbt2o6uY5Ht3r0bbdu2tZqvbdu22Llzp/Ta2k2eQ4cOoWnTgk9GPi2cPL3RctpMqbWILa1EouMfICItAxkufjhZvSM0SifphpwAQKlQYHC/HhDygga64IGAM+cu4Mh/J3V5Bd2NXGk5QQ0BaqkM3TygacO6aNaovi6QIROksuYuWIb0zMx8ZTyqQ/55ADDxjdFwzrdtp89exJ+bttm1z+rXKYuuUQWPld8Wr8K9OHNPcAomp8cOb26my5WbZtd/7tJVnBCqQy1Yv1mhFFXoVS4XKlFAjkZAjhbI1eqnBeTmzcvRCLr5WgGC0hnp2WpotI7tikorikjOyHvqVmZ5QGtDMoh4Z815BG67JbU68XF3hoezApqg6nCS53UBZtgNmFyE3MRH09PdreBMAIEBfqhetZJd29OomekuV6Z/PxtCcjzCzu6DMjsN08uEWi0rUy5D9N1EOLm6wb9WfTh5eiEpIwvX78YX6E5KJVcgR5bXtZRcaZQe0SgCL7SPLHBestTlir5nJ8OjSRQBZVglpMiUumBGYgZSMh/i0o0UqIOq5x0vkI6Z3LzXOVoBmnytmC5c1WLjVeOxpXQUgOxxdMiiOyacZCKc8o4RXw9X1K1eEd7uzvBybYjVf2hwE8Fw0qrxcvxfKK0y7jrpjjIAvwV1R5bcBUpRhb4VciETdAGUfr0aomqlCkb509LS8WPaBdjDz9cbHeuXLzA/8WYIcpJKF1zAAnNP0dt7jDspTZ9j/Hy97S4rMMDP5PwqFcshLcO2Vjp6MqHgOBpenp521ykkONDk/ArlwuBipsWGOab2lbubq911Cgs13aovLDQEWWZabJjj7uZaYJ6TUml3nUqZ2U8hwYF2l+XlWbBljkyQ2V2OI8/lfr6mBw+vWrmC3d1nmeLj7WV3nYICA0zOr1ihDHx87HsoQiEv2IWjh4cbqlethAuXrsBwCCt9YEQ/TUREREREVJIVaqD1Z8358+fx5Zdfmk1ft24dUlNTMWXKFEyePFmaLwgC3N3d0adPH5PLffLJJ6hUyb6LYUOPc6D1xcN6Y16Il02DDQ++n4MR85fC3cfXYl5yvOHjJuKm1h935SG6u9emAnN588PEOGz/+X2716HVikjPzkVKZi6SM3KQnJ6NZP10RjZSMgym8+anZOQgNSvXAVvoGK5OCmlMFG+3R+OjeEvznODj4aKb7+YEH3cXeLk5QSF3zMDFuWkp+HBsf6wNtO0GVnRCKj6fvQJOnqZv0BVGjkqta5mRkYPUTH13U7r3KSVD32ojR5pOzcxBcmYu0jJzoX3KTvtOChl83F3g7eYEbzdneLs7wctN/97qp3X/9e+vl5szPF2VVgejNvxMuaqz8HLC31JgRAqIyJyL9JkiInraDB83EWp1we61TFEo5Bxo/TnDgdaJiIiIiCwrEWMrVq9eHQsXLjSZduvWLSxevBiCIGDo0KEF0gMCAswu+6xw8vTGnyGVEeecbDVvnLMTVoRUwusMiDwxIdp4ZAhuSDHXjZYgwFubilDBhsGCTZDJBHi5OcPLzRllAgo+2WuOWqNFamYukvNuvuuDJfmnU/KlZ+aqC1VPS7Jy1cjKVeNeUob1zAY8XZXwcXd5FDhxd4G3u1NeUMUlL4DiDB8Pl7x0Z3i6OhVonSF398QWfz8Atm3bFn8/fO1ecF9rtSLSsnN1+9UguJE/qJGalSvt29S8QEi2yrabXY+LIABerrqghpebE7zzAlbeBtNebk7wMZHu4lS8X0XSZ0rhhd+CuuPl+L8AQGohAqBInykiIiIiIiIiInp2FOpO1Pbt29GhQ4cir9xR5RTF4sWLodVqERkZiQoVKlhf4Bmk0WpxVWn7zeMk5X1otVrIrDyBTcVDBqCS5jrui0GIl/kbdaWlEFUI0iYiRBsPmaJgtxnFSSGXwc/TBX6elsfQyC9HpUZyRm6BYEn+AEr+IItKY/t4I7ZKy1IhLUuFWw/SbF5GLtONl2LYKiVNdQsZgsbmsSoyBA0G/LgQnsowqXVHStbT2WrDWSmXWmt4u+b9zwui+bibD3p4ujhBJns6x5XI/5n6Lag7ACBL7vJEP1NERERERERERPT4FSooEhUVhdatW+Pjjz9G+/bt7V5+69at+Oyzz7B//35oNE/2aefff/8dADBs2LAnWo/itPXUUWQJKgh5915l+UeozqOFDICILEGFf04dRee6TR5fJQkAoJA/+kiWRSLCkIg00Q0ayCGHBp7IhEwGQCY3yvs0c1YqEOyjQLCP6X7iTRFFEZk5aqPAiWHAxFxrldRiCDJotCKS0nOQlG7Y578AZ3QCALhqsq2OVQEAJy4DwF2H1s0cw1Yb3u55QYy8gI5+Wt9qwzD9cbTaeNxMfqZkeZ8p8dn8TBERWWPP+YznPiIiIiIiKmkKNabITz/9hKlTp+Lhw4coXbo0Bg0ahDZt2qBx48bw9S3Y7dLDhw9x5MgR7Ny5E8uXL8fdu3fh5+eHKVOm4LXXXnPIhhTG4cOH0bRpU7i5ueH+/fvwzDdIqCAICAgIwPjx43Hjxg24uLigVq1a6NmzJ4KDg4u8/ifV329uWgr2ffw6Uq9eBAAoS1dE2y9/5hgi9MzTd0eVv1uvJDNde+mn07OLNuBu/sBI/oBIYbko5brxNNwfBS30gQ5vt2ez1QYREREVL44pQkRERERkWaEHWk9JScFXX32FX3/9FQ8ePJD63A8ICICfnx88PT2RmpqKhw8fIjExEYDu6e+AgACMHj0a7733Hry9HTfocGG89tpr+PnnnzF48GAsXry4QHr+cQT0XFxc8Pnnn2P8+PE2rUd/YZLf6dOnUaNGDezcudP2SjuIKj0V/302EQAQ8dF0KD1sGzCa6Hmk1miRmpU3tkdmbt7A5LmPxvXIVOX9fzT/YbpxF1/6wAgAkwERfw9nBPu4wdvNCZ6uTnn/lXktOvTzdK+98l67KNmdExERPT88PDyedBWeK0/jNQYRERER0eNU2GuMQgdF9HJzc7FmzRr89ddf2Lt3L+7du1cgT2hoKCIjI9G9e3dER0fDycmpKKt0iNzcXJQqVQoPHz7Etm3bTI5tMnToUAwYMAB16tSBr68vrly5grlz52L27NnQaDSYM2cOXnnlFavrelovWFTpqQDAgAhRIfx7JR6vzt1rNM9Vkw0AJluIzBkTiYaVgh5L3YiIiJ5GDIo41tN6jUFERERE9Lg8saBIfg8ePEB8fDySk5Ph4+ODoKAgBAQEOHIVDrF27Vr06dMHYWFhuHHjhl2Dis+ZMwdjx46Fn58f7t69C2dn50LVgU3biZ5duWoN2n2yBg9Ss6zmDfByxa5pfaGU236eISIiIioMXmMQEREREVnm8Dt0AQEBqFmzJpo3b46aNWs+lQER4NEA60OGDLErIAIAY8aMQWBgIB4+fIiYmJjiqB4RPeWcFHIMjKxuU95BkdUZECEiIiIiIiIiInoKlMi7dAkJCdi8eTMAYNiwYXYvL5PJULlyZQDA3bt3HVo3Inp2jI6qjdbhYRbztA4Pw6io2o+pRkRERERERERERGRJiQyKLF++HCqVCk2bNkW1atUKVUZSUhIAwN3d3ZFVI6JniFIuw09j2uGtbvUR4OVqlBbg5Yq3utXHT2PasZUIERERERERERHRU0LxpCvwJCxatAhA4VqJALrBC8+fPw8AiIiIcFi9iOjZo5TLMLZzXYyKqo1jV+KQkpkLbzcnNKgUzGAIERERERERERHRU6bEBUXOnDmDY8eOwdnZGS+99JLZfIsWLULt2rXRoEEDo/nHjh3DwIEDAQDR0dEoXbp0sdaXiJ4NSrkMTaqWetLVICIiIiIiIiIiIgtKXFBE30qkR48e8PX1NZtv3bp1GD58OKpVq4aaNWvCyckJly9fRmxsLLRaLSIiIjB37tzHVW0iIiIiIiIiIiIiIiqiEhUU0Wg0WLp0KQDrXWcNHToUbm5uiI2Nxe7du5GWlgYvLy9ERkaiX79+ePnll+Hk5PQ4qk1ERERERERERERERA5QooIicrkcd+7csSlv79690bt372KuERERERERERERERERPS4OGQX4999/R05OjiOKIiIiIiIiIiIiIiIiKhYOCYoMHz4cYWFhmDBhAi5evOiIIomIiIiIiIiIiIiIiBzKIUGR6OhopKam4vvvv0eNGjXQvn17rF69Gmq12hHFExERERERERERERERFZlDgiKrV6/GrVu38Nlnn6FcuXLYtWsX+vfvj7CwMHz44Ye4fv26I1ZDRERERERERERERERUaA4JigBAUFAQPvjgA1y9ehWbN29Gz5498fDhQ3z55ZeoXLkyunTpgg0bNkCr1TpqlURERERERERERERERDZzWFDEUMeOHfHHH3/g5s2b+PTTT1GmTBn8888/6NWrF8qVK4f//e9/iIuLK45VExERERERERERERERmVQsQRG9kJAQtGjRAo0aNYIoihBFEXfu3MGUKVNQoUIFfPDBB9BoNMVZBSIiIiIiIiIiIiIiIgDFFBRJSkoyGnR9zZo1qFy5Mr755hscP34ckyZNgqurK77++mt8+OGHxVEFIiIiIiIiIiIiIiIiI4IoiqKjCjtw4ADmzJmDtWvXIicnBzKZDF27dsW4ceMQFRVllDc+Ph4RERHQarW4c+eOo6rwTGnWrBkA4NChQ0+4JkRERERE9DzgNQYRERERkWUKRxQyc+ZMzJ07F2fPnoUoiggODsbLL7+MV155BWXKlDG5TFBQENq1a4clS5Y4ogpEREREREREREREREQWOSQo8tZbbwEAWrZsiXHjxiE6OhpKpdLqcrVr10ZkZKQjqkBERERERERERERERGSRQ7rPGjduHMaNG4fw8HBH1KnEYNN2IiIiIiJyJF5jEBERERFZ5pCWIj///LMjiiEiIiIiIiIiIiIiIio2MkcUkpOTg5s3byItLc1snrS0NNy8eRO5ubmOWCUREREREREREREREZFdHBIU+f7771GhQgXExsaazRMbG4sKFSpg5syZjlglERERERERERERERGRXRwSFNmwYQPKlStncdD0yMhIlC1bFn/++acjVklERERERERERERERGQXhwRFLl++jBo1aljNV7NmTVy+fNkRqyQiIiIiIiIiIiIiIrKLQ4Iiqamp8Pb2tprP29sbSUlJjlglERERERERERERERGRXRwSFAkJCcHp06et5jt9+jQCAgIcsUoiIiIiIiIiIiIiIiK7OCQo0qpVK5w5c8bieCEbNmzA6dOnLY47QkREREREREREREREVFwcEhSZMGECZDIZBg0ahOnTpyMhIUFKe/DgAb755hsMHDgQcrkc77zzjiNWSUREREREREREREREZBeHBEXq1auHmTNnIicnB5MmTUJISAj8/Pzg5+eH4OBgvP/++8jOzsaPP/6Ihg0bOmKVREREREREREREREREdnFIUAQAXn31VezduxedO3eGq6srkpOTkZycDFdXV3Tp0gV79uzBuHHjHLU6IiIiIiIiIiIiIiIiuygcWVjz5s3x999/Q6vV4sGDBwCAgIAAyGQOi70QEREREREREREREREVikODInoymQxBQUHFUTQREREREREREREREVGhsAkHERERERERERERERGVCA5tKXLkyBGsXbsWFy9eRGpqKkRRLJBHEATs2LHDkaslIiIiIiIiIiIiIiKyymFBkTfeeAM///yzFAgRBMEoKKJ/LQiCo1ZJRERERERERERERERkM4d0n7Vo0SLMmjULdevWxdatW9G7d28AwIULF7Bx40YMGDAAAPDBBx/g6tWrjlglERERERERERERERGRXRzSUuTXX3+Fi4sLtmzZgqCgICxduhQAUKVKFVSpUgWdO3dGhw4dMGrUKERGRqJcuXKOWC0REREREREREREREZHNHNJS5PTp02jevDmCgoIAQOoiy7D7rBEjRqBWrVr4+uuvHbFKIiIiIiIiIiIiIiIiuzgkKJKdnY3g4GDptaurKwAgJSXFKF/t2rURGxvriFUSERERERERERERERHZxSFBkVKlSiE+Pl56HRoaCgA4f/68Ub779+9Dq9U6YpVERERERERERERERER2cUhQpHr16rh06ZL0ulWrVhBFEV9//bXUhdaePXuwZ88eVK9e3RGrJCIiIiIiIiIiIiIisotDgiJdunTBzZs3ERMTAwCIjIxE48aNsX79eoSGhiIiIgIvvPACRFHEhAkTHLFKIiIiIiIiIiIiIiIiuzgkKDJ48GD89ddfRuOKrF+/Hp07d0ZCQgJiY2Ph4eGBr776Cn379nXEKomIiIiIiIiIiIiIiOyicEQhPj4+6Nq1q9G84ODg/2fvvuOjqhL+j3/v1PRCDb2DSBVUwEZRRNG1s5a1d1197GXVfdZnF/W3XdcVXXvBhmvvKIKKBKV3SCgJNdSQnky59/fHFFImkAkTEpjP+/Wa18yce+85Z2aSSe5855yjzz77TOXl5SoqKlL79u1ls8UkgwEAAAAAAAAAAIhaTEKRu+++WxkZGfrf//3fOtuSkpKUlJQUi2YAAAAAAAAAAAAaLSZDN/79739r6dKlsagKAAAAAAAAAACgScQkFOnQoYMMw4hFVQAAAAAAAAAAAE0iJqHImWeeqR9//FEVFRWxqA4AAAAAAAAAACDmYhKK/OlPf1JCQoIuvvhibdmyJRZVAgAAAAAAAAAAxFRMFlp/4IEHNGjQIH322Wf65ptvdMwxx6hbt25KTEyss69hGHrppZdi0SwAAAAAAAAAAECDGZZlWQdbic3W8AEnhmHI7/cfbJNHhFGjRkmSsrOzm7knAAAAAI4EnGMAAAAA+xeTkSIzZ86MRTUAAAAAAAAAAABNJiahyOjRo2NRDQAAAAAAAAAAQJOJyULrAAAAAAAAAAAALR2hCAAAAAAAAAAAiAsxmT5r3LhxDd7XMAzNmDEjFs0CAAAAAAAAAAA0WExCkVmzZh1wH8MwZFmWDMOIRZMAAAAAAAAAAABRiUkosmHDhojlpmlq48aN+uqrr/TUU0/prrvu0o033hiLJgEAAAAAAAAAAKISk1CkW7du9W7r0aOHRo8erZNPPlnnnHOOTjzxxP3uDwAAAAAAAAAA0BQO2ULrEydO1JAhQ/T4448fqiYBAAAAAAAAAADCDlkoIkm9evXSkiVLDmWTAAAAAAAAAAAAkg5xKLJy5UrZ7fZD2SQAAAAAAAAAAICkQxSK7N69W7fddptWrVqlE0444VA0CQAAAAAAAAAAUENMFlrv2bNnvdtKSkq0Z88eWZal5ORk1hQBAAAAAAAAAADNIiahSF5eXr3bnE6nunbtqrFjx+rBBx9U3759Y9EkAAAAAAAAAABAVGISipimGYtqAAAAAAAAAAAAmswhXWgdAAAAAAAAAACgucRkpAgAAAAAAAAAADjyeXx+LVq/Q0XlHqUnuTSsV3s57YfP+IuY9PSLL77QuHHjNHPmzHr3+e677zRu3DhNnz49Fk0CAAAAAAAAAIBDxOs3NeXLJRr3+//qqqe+1v+8MFNXPfW1xj7ynqZ8uURe/+GxzEZMQpGXXnpJCxYs0PHHH1/vPiNGjND8+fP18ssvx6JJAAAAAAAAAABwCHj9pm57/jv967NF2lVcUWPbruIK/euzRbrt+e8Oi2AkJqHIwoULNWTIECUnJ9e7T3JysoYOHapffvklFk0CAAAAAAAAAIBD4IXpy/T98s373ef75Zv14vRlh6hHjReTNUUKCgp0wgknHHC/Ll26aP78+bFoEgAAAAAAAAAAVGNZljw+v8qrfKrw+FTp8anc41NF8H74UhUorwzeD+1fUeVThbfa/lU+lXu82lZYVqOdRH+lJKnCnlCj/M0fVuv60we16DVGYhKKJCUlaceOHQfcb8eOHXK5XLFoEgAAAAAAAACAw47H568TRtQOL8qrfKr07gsvKoL36w8vvKrw+lVR5ZNpWU3a/0R/pa7b8akk6aV2v6oRjOwqrtDCdds1om+HJu3DwYhJKDJkyBD99NNP2rp1qzp27Bhxn61bt2r27Nk67rjjYtEkAAAAAKCF8ZQUSZJcqenN3BPgyMDvFAA0D79pNngkxYHDC29wdIZfFR6vKqp88plNG1o0pUR/pa7Z8ak2JFVJkq7d8alerhWMFJV7mqt7DRKTUOSKK67QrFmzdN555+nDDz9Up06damzfsmWLzj//fHk8Hl1xxRWxaBIAAAAA0IJ4Soo0+5HbJUknTX6aD3GBg8TvFADUzzStwOiKWmFE7Wmfqk8XFQgpvKr0+IP71w0rQvt6fC1/sfBYcjlsSnQ7lei0K9HtUKLLqSSXQwkuuxLdTpVXevXT6q3hESI73OX6oG07SdKVBTt03Y5Pa4wYSU9q2bNFGZZ18GNpTNPUhAkTNGPGDLndbp166qnq27evJCknJ0czZsxQVVWVxo4dq2+++UY2W8udT+xQGjVqlCQpOzu7mXsCAAAA4EjQXOcYoQ9vi9avkSSl9+zHh7jAQeB3CsDhzrIsVXn9NUZSlNcTVoTWrKj0hPb3BqeT8qu8VlhRER6J4W/uh3hIOWxGOKxIdDuC4YVTiS6HktwOJbocSnA5lORyBPerdnE7ggGHo8b+oUuCyyHHAdb/8Pj8mvi7N3TeuvfUwbtLf+3SUdvdgeCjfZVH923aqm3ONnqp3a+UnJmpmZMnteg1RWISikhSZWWl7rzzTr388svy+Xw1tjkcDl177bX65z//qcTExFg0d0QgFAEAAAAQS4fyHOP6238nn98np9+rE3YuU5q3VHPSUyVJJxSVqNiZojltB8lrd8phd+jFp59o8j4hstBr1RC8Vs2n9u9Uhre0xva9/E61CPw+4UhhWZa8PjM8sqK+hbgrPN7gSIrAyIryYFhR4fHud2HuCq9PTbysRYtiM4w64UN9YUTt8CK8T3D/8D6h8MLtbPaA4ZZb79XQgmXK8BVpQUqy3mvfpsb2Kwt2aGhpubY42+jr9ifo02cfbqaeNkxMps+SpISEBD333HN69NFHNWvWLG3cuFGS1LVrV40dO1bt27ePVVONcvXVV+u1116rd/tNN92k5557rk75unXr9Oijj+rbb79VYWGhOnfurIsuukiPPPKIUlJSmrLLAAAAANBi+fw+GZ5Kjdq9Qhm+Mi1KSdIHbVtLklL8fg0tLdWxO1dqTqt+8rkSDlAbmpLP75PPF1/fqD0c1f6dMqUaQWOGt1SjdizVj60H8DvVjPh9wqHk85v7WYi7ZljRmIW5/YfxuhaNUd/IiQONpKg3vHA7leC0K8ntkMthl2EYh/wxBUbkeFXp9ajCWxW49lSpyutRRfB2pdejSo+n1j6ewD7B7RXeqjr7hO97KlXmLNNX3TIkZUTsx9eZGRpcWq5O3l26YPt38pTc1qJHN8YsFAnJysrSJZdcEutqY2bChAnKysqqUx76RlV1Cxcu1JgxY1RSUqJhw4bplFNO0c8//6w///nP+vzzzzV79mylp7fcFxcAAAAAmorT79Wo3SuUGfzwdnpmRnhb6MS4nbdII/fkan7bo5qtn8DhovrvlCQtqRM0livTV6aTd69QdrvBzdlV1OI0vZIkr83ZzD3BoeY3TVV4/LUW4q67ZkXEsCJ0TH3hhccnrz++1rVwO+0NCyNqj6SovX9wWqkEl11JwemmEpyHLrSwLEsev0+Vnn0BRChsqAreD4cV3vr3qTfg8O7bVuk9RAua2/b/3G13u7Q0JUlDS8uV4S3V7Edub9HTPsY8FGnpHnzwQY0ZM+aA+/n9fl166aUqKSnRE088oQcffFCS5PF4dOGFF+qzzz7T/fffr//85z9N3GMAAAAAaFk8JUWB6X2qfXgbmldaqnlinOUt1IidK+UpKWqxJ8bxiA9xW5bav1P1BY02SZm+Mp2wcxm/Uy2E0/Tq5N0rJEk/th7A71QLY1mWKr3+8JoVobUoQmtWRFqIu3ZYUXc6qX2Xqjhb18Jpt9UKHxxKcNYNK/aNpAhsrz4NVHjxbpezxnRSiS677E24DrVlWfL4fMEREFWRR01UDyjC+1QLIKoqVVFVqQpvlSqqqgKBhsejSl8o0PCqyudVpc+r+BqDE1D9b1XR+jUtOhiJSSjy9NNP684779Qnn3yis846K+I+n3/+uc455xw988wzuvnmm2PRbJP6+OOPlZOTo4EDB+qBBx4Il7tcLj3//PPq2rWrXn75ZT3++ONq3bp1M/YUAAAAAA6d0ALQofUOan94G1L9xLitt0j/veFi5Q8eL3+taX9cTqfuv+PGOsfPnbdI334/J6q+nXLCcTrlhOPrlD/17CsqKSuPqq77br9B7mpBjyQtX5mjj774Jqp6jhl8tM46fWyd8pfemKZt23dGVdct116m1q0ya5Tlb9qiN979aL/H+f01P7Tb34e4fr9fk//2TIP7dMXF56lbl041ynbvKdSzL7/V4DokqUP7trruil/XKf98+kwtWroyqrrOmzheA4/uW6Osqsqjvz79QlT1pCYn6Y5brqlT/sOcX/TDnHmRD7JM2Xxe2f3e4LVHNp9XQ/r2UPesNvKWl8lXXiZveZm85aVatWyFVFWhpNLdyjD3vU77CxolKcNbqq+uOUcZfY6WOzVNRZVebdyxW6bdKb/DKdPuCl47g2UumcH7frtTstk04tihGj/mxDoPYcpLU7WnsCiq5+qOm65SamrNKcZz1m3QtA+/iKqeAUf10flnn16n/I13P1T+pq1R1XXt5ZPUMatdjbKtBTv08tT3oqqnW5eOuuLi8yNuC/0uhUb3nLhnlX5s1V/+4O9U7d+nX58/UX179ahRR0lJqZ76T/3TvUfSKjNdt153eZ3yb2b9pJ/nL46qronjx2jYkAF1yqN5H5Aa/15uWZJpST5L8lmGfKY0eNAAHXVUv31TRQXDiOnfZ6us0hvY1zTksyR/8Bhf9eta2+OJ3WbIYUg2+WU3JIfNksNQ4BK+bclhk+zBa4dh6aTjh6pzVlslBsOKRLdDnopyffDJFzWOjThYwJJUGbh0aN9W113WwPdyy5JkyfL75Te98pu+ahe/hgw8Sq3apKuyqkoVVZWq9FSqtKJCC5atkM/yy2eZgYtM+S1LXpnyWZZ8Cly8suST5DMsmTabqixT8TXuZv9sliWnZclpWnJZZvi207LkCt8OlLuC5budDi1NSa63ztp/q1qymIQiH3zwgbKysjRx4sR69znzzDPVvn17/fe//z0sQpFPP/1UkjRp0qQ6Q6s6dOigk08+WTNnztTnn3+uK6+8sjm6CAAAAACHVCgQKVq/JlxW+8PbkO1ul6Z0zNKw0lJ1r6xSVlmh+mVPk082+Wx2+YzAxbQ7NWfvajkSE+VISJIjMUmOxETt2FQgz/I1gf2q7e8zbDXKTGPft0r79+0Vsd+56/O1t6g4qsdqWnU/OikuKdHqnHVR1dOhfduI5RvyNytv4+ao6vJ4vXXKysoroupT7Q9xT969okYwYlmKqr6y8oqI/Yz2eaqsrIpYXrB9Z9R1FZ88sk6ZaZn7r8ey5LD8clp+OU2fHJZf3kS3Nv/YORBklJUGrivKtGv1GmVs3CSH6Qvu75fDCtx2RPi5kaS9y6XFEcojrVTqk/RFrfBLkj5o01rlNpuS/KaSTFNJflM7Vy9Vkt8vt2Upmq9r+mRTyc/va/rbreRMSpYjKVnOxGQ5klNkLVkle6VX3uDvWODaIW/4tl1ewyGvzS4r+Pvn89f9tnxpaXnUr11mRlrE8vxNW6Ouq6qq7s9UVVVV1PXUx256NXLPmvDvkiS19pbo+D3r9FmbkcrUXtlq/T6VRvig0Of3R92nrHaR31d27NwVdV0nHD8sYnn1ekxJpmw1Ln6j5n2706V2360ITw8VGm2xLn+r1m/yy5QR2Dd4nL/asar1udunWzZIX22I0CtDUt2/N4cVy1KCy6HkBOe+ERdOh0y/V5s2bpJNpuwyZbNqP+Om7LXKju7dXTdcOSk8tVSSyyGHYenJZ17U0mUrZZMVqMeywrcNWcH7wfLg7WGDUtSxbK/MIq9Mn1der1c7d++SY/ks+WXKa1gyZck0LJmG5Jclv2HJL8lvBIIHvyGtznHo58Wfqsr0q8o05bEsVVmmKv1+VVmWfDZDXsOQ15C8hk1emyF/PVNavZ+d2/Dn1ah1Xed5PzziECMYVIRCCKdlBkMLq1ZoYR4wwAjsb4aPrV6ny7Rkj7JvpqS/dul4wP1CX4rJ7NmvxY4SkWIUiqxZs0aDBg3a77xsNptNgwYN0ooVK2LRZKN98MEH+uCDD+TxeNSlSxeNHz9exx9f95tEixcvliQNHz48Yj3Dhg3TzJkztWTJkqbsLgAAAAC0WPWNEglZn5Sg9UmBkSEJflPdKqvUvbJSPSqr1LWySqmWJfmk7fN/inj8MQ3qgxH+4NZ4L0ezfpgWCFcSEoMBS5J671itUo9P/mof6oYvtmpBS/BS+wO6I0XtQERSeI2Kw3ban2CY4QiGE07Lr6p1K7XFu7faqIxSVRQXadje3Gohhl9OyydH8BinFXkKnHl/mV+nzJDUqe6ujVZusyk/wa38BLfyEtzakOCWN8IUMqUOu/7brk3EOmyWpSS/qUTTVJLp33c7HKD4w0FKkmkq0W8qyVshc+umOh+MdQxeGsIvQ16bQ3PvuVrulNR9AUtSsgrLqzSwOE9eWyBQ8VUPVWyOYLgSKIv0wXhLFghEctXOW3dETUfvbp2162e913aMOlnb1HSTAe3jN01VVPlUUmWqUq4a4UONAMIwapYZNr2SvVUfrP6u2loWXlV4/drq6B/e12rIa2NJ/+/9ekZQ2SKHXS1VjUDCCoSebVMT1atTOyU6pESHlOQwlGCTli5ZosqSYjksX/i9yK7Ae4o9eHHIL7tlyiG/bJapiaeerGS3S5bfJ9Pnk+nzaveu3Vqye3mwbStCkGHKsAKhhE+BYEJFczV9+Qeq8vuCAYQpjyx5DZvaGIa8NkMeIxBCVNpswTDCkCcUTATLPIahmd+/L68ttN22L6hoZ0jRfnxuBte4CP/w2yTHofhNaDrhcCFCKHGgECPyKAyzxrFuu11uu0OGJVV4vDINmywZ+35nDSMQLBq1bsumjh07qFfvnrI5HLI5nLI5nbI5HJr+fbaKyypUHvG4WnUHb993583KyMyUzemQYQ/Ut3zNOj3/4utKr9oQ8UswtW13u5Sd3kqPt+BARIpRKLJnzx61aRP5D3N1bdq00a5du2LRZKM9/fTTNe4/8sgjmjhxot544w21atUqXJ6fny9J6ty5c8R6QuWh/fYn0iLukrR8+XL1799fpaWlDeo7AAAAcLhKSYn0fWg0VrOdYxh2HfO7/6cFj92nkry19Y4SiaTSbtOa5EStSU4MVGVZ6ljlUffKKvUIhiWZPn+9X/Ksj02WXJZPLr9P2rtThXvrTknVLco6TZtdM647V46ERNkTAwGLPSFR5V6/xvr3Bqckcshvd4RvB6YlcgRu25wyHQ75bU5lpiSqpKSkzpcIO3dsL6czulNyr8db57U1ZKlPr+77PW7tujw5IgQiIdWDEZ/Nqd4HqK92+7X75PV4D9gnWZYM0y+73yu7z6M2CZbys7+Xr7xMvopy+SsCU0xl5uRqjLkjMA2V3yt7aGqq4G2b31vnZ2bv20v1S4Qme0QoO9RMSdvdbm1KTVF+YqLW2g3tdET7fd0I9RqGSh12lcouKbpwyx0OS+qGKYm1wpTq+7ksS3ZZspteVe7Yqsoddevu18A+WIYhv90p+4wV+mbxl+FA05GYLHtiknoVFKi10xOeEsxvd8p0uALXwSnD/HanLNu+UNPv99f52fT7/Qf+2aylY/u2Nepx+r06ds+aiIFISCfvLk3aOUtftBmhEb0yZFmS35JKq3xas3G7KjyBdS0qPT7tKSqTK6u3fJYhf3gKqQhTQ4W2m4bWlts1+qF3A2tleHzy+Kp9C955VFSPb9fGEmljSd0NRhOFpMGRCXYrMCIiEBwEbjuCt23Vbtstf539nJZfTlW/mHLKL0fwOEfwtt3yK6tVmhyygvcDIYe3skLlJSXBURT+4LUZDB5MGWag3AgGEWGr6z6crlI4bPDWCCJsNUIJr81QuWGTx27ole8/rbavLXzbm2HIY9jlqRZgVK/D14TrbRyOHMFgof6wQnJagQ+/k51OuQxDbsMml80ml80uy+uTr9IjpwzZZchp2GSXTQ7DkMOwySGb7IZNdptNlmGXZbOpe7eu6tK1s2wOhwy7MxhEOPT199ny+i1ZNpssw1b32rDLshmyDLuqDJsqbTb9z81Xy3A4Zdhs4f8Rlq5YrW9mRTd1aK+hA9X3xLpf+LeVJsgKjpK1SQ0KaF1ZnWULTo9lSfJLcqWmqWPffvrv1r3BkgP7qnUr/a9lyHMIPvNu7DmGYVnWQa/70qlTJ3Xu3Fk///zzfvcbMWKE8vPzVVBQcLBNRu3JJ5+U2+3WuHHj1KVLF+3cuVOzZs3SQw89pK1bt+qEE07Qjz/+KFvwDcblcsnr9So3N1e9e/euU9+LL76oG264Qaeffrq+/vrr/bZ9oBOW77777uAfIAAAANCCEYrEVnOfY3hLi/XBrVfohay0BociDZHm86t7lUc9KyrUvaJSnao8UU/v0CIZtmDAEghXHAlJsickVLsdKk8M3k6SIyFB9sSkeve3ORoeqNx1zyMatWNpxECkukJHsrLbDdY//z653n0sy5Lp9YTDC195mfwV5fJVlO0rqyiXr7x0v9v9FWWyzMNjOpPqQh/SOxKT5EhKliMxOXzbHvwQP1CeJEdSksrtdq0pLdLKvTu1Yuc2rdy+WaVVlc39MGIiNDolFJKEQpPE+kamVAtdmur32pIhn8MdvnjDtxP2ldsjlNU4JiGw7opR9yNEp6dcg2b/R1newnCZKWlOeqok6YSikhofPG5xttEbHc9VkdW0Uz4ZVqSQITCd275QIXLA4AiGAvYDBBGh26FRD3W2VWvDITN8CfXFbpmymYGRE00t9GFuKFwIj4wIjo7w1BNghEZNBG5HOrZmiOEJ1od9HJJcMuQyQhdbOIhw2+yBERG2wKgIt8OhBLtDbodTbodTCQ6nEpwuuZ0uJYauXW4luhOU4HIr0Z2oBLdbSQlJcrvccjhdMkKjIxwOGaFru2O/sxkhet+smK973n5WUuDLKPtjBr+q8PdLb9H4Acc2ed8ae44Rk5EiI0eO1EcffaQ5c+bohBNOiLjPnDlzNG/ePJ1zzjmxaDJqd955Z4373bp101VXXaXTTjtNgwcP1pw5c/T+++9r0qRJMW87Ozs7YnnoRIYTRAAAAADRaPZzjJQUfZTVW9vdexu0e5KvjZLT07SrdJsso+4aFCHFDruWOhK1NDiaxGbY1Tm1nY7J6qyTu3bVwIw2SjZN+SpDH6xXhG/7KioCH7xXVgQuoQ/nK8rlq6yQ6Ym8ZsUhYZmBvlXsP5SIhs3pqjFFmD0hUc7EpH0f2Ae3GTabxm5foES/54B1ZvrKNK5gvta8+KQsv1feautoeMtKA1NRVZTJ8vli9jgOFXtCopxJKeGpnULTPAXup1Sb9mnfPo7klMA6G0nJciYny5EQeD7r4zdN5RZs1sK8HC3Kz9HC5blavyO6xcFtliXzAB/mpXtN+cxjVGWzSYZXlrz7ruWVZdS8ltE0r1fN0SnRSQiGJIk1RqCYSgwFLKHyUMASDFNclrXf0WSGLDl9lXL6Di54MiV5DJcqbU5V2VyqMlzyGHZ18uxSolXzd2lJSpI+aBtY0SXF76+xwHAn7y7dsvkdzUvuL9Ow1RtQHDi8CO0TWlsieD8UNhzgQ8qWIBRUVEQIHELhQp3pnQzbfkOJ2iFG9X0bNOVXnHAYNrntdiU4HHLbA+GD2+FQgtOlhGAAkeB0BYKHYAiR6EpQYkKCEt2JSnQnKMmdEA4pwse4XEpwusNlCU6X7IRER6TzR4zRx699Lp/PH3EqTinwxYrQVJwOh13njxjTPJ1toJiEIrfddps+/PBDnXvuuZoyZYouuuiicCJnWZbef/99/fa3v5VhGLrtttti0WTMdOrUSddcc43+/ve/64svvgiHIikpKSosLFRZWeR/WkNDJ1NTUw9ZXwEAAACgJfCbptY7o/iA31akuX98Wj7T0vcr1+mLRYu0cEOOthRtlt8qkozIH+iZll8bi7dpY/E2fZwTmKu+VXJrDenSR6cOHKyRffqrb9sODfpGqOnzyVdZIX9lubyhQKWiPBiqVAtUaoUp1QMYb0W5/OH9y2WZTf+N53ofj9cjj9cjT0n9U/iEJEZRb4Lp1cYZnza+YzFmd7kD4URwVIYzOSW8KLgzuQHBRlKKnIlJMuyxH5tQWFaixflrAyFIXo6WbFyn0qr6Q7/abLJLVroMM0OGmSHLqJLfceB1WIucNjl8huxmlhryWbglU5KvZngSCk7kCZb5JMNTJ2Cp73fzYFXabaq026Kd6Ut2y6ozCiUQmPhrjUqpOVol0TSjWtvDJinB8ijB75H89b/X1V5XKbTAcPW20v1lOq247to0LYVfqj9wCIUN1UKJ0JoTkQKM2tNFBUKLfaMwDhT4xRO7zaZEpzsYNITCiVr3q4UNte+HwwmXW26nU4lOd819XK5wmaMJ3v8Qv7w2p35sPaBGMFI9EDlcxCQUGTt2rH73u9/piSee0CWXXKLU1FT17NlTkrR+/XqVlJTIsizdf//9Ou2002LRZEz17dtXkrR1675vcHTr1k2FhYXavHmzhgwZUueYzZs3h/cDAAAAgHgyfdk8VRje8Oel9X1LOfAxpKUKw6uvl83TmUNGaPzgvho/OHAO5vObWpK3XZ8tXKS5uau0fle+POae4Ae2ke0p262Zq3dr5uq5kqQER6KO6thDo/sP1Al9jtbgLr2U4Ko7VY3N4ZArJVVKSY0qJKhPeBqp0CiVyvK6gUpoqqjKaqFLeTCIiRDA+Csb/oF6S2dzuuqMyqgTVkQodyYny5GYErxOjmqasKbkN02t2bZRi/NztTAvVwvzcrRh57ao6rBZSZKZIZuVIcPKkGGlygh+fG7Jktc+u+H9sa1VnzZ9lZrk3u9+BzNhumlZMi2f/JZHPrNKPtMjv+mRzwrdrpLPCl7X2mZa9f8OHwy/YajE4VCEFTAOKCHiovORw5Tq66q49vMk1l5XabvbpaUpSTVGizSGKdUKHSIFDrbwKIl95bZaoydqhhKRQgw/QUWYYRg1R01UCyD2jZJwVgsvqo+aCIUS9QQcTne43gSnS067nSmecFhx2Pf9PbZkV3a7wTph5zJJUnbbQbLsznDQUH3flipmPXzsscc0aNAgTZ48WStXrtTixYvD2wYMGKCHH35Yl1xySayai6nCwsBckMnJyeGyoUOHavHixVqwYIHOOuusOscsXLhQkiIGJgAAAABwJDtzyAitf/Ld8H1PSZF+fOQ2Fa/PkSQ5O/XU2P83RckZmfutx2G3aXivDhreq4OkiTJNS2sLCvXlohX6ftVy5RTkqdy3U5ZR/ze1K30VWrxxpRZvXKmnvpZshk3dWnfWCX3768S+AzS8e1+1S99/PxrDMAzZXW7ZXW65Y1S/ZZrhcCQwmqV832iW6oFKrdCldsDir6yQt7xU3tLGfHQcYHM4ldypq1wpaXLWGKmRfMBgw5GULLuzaddQaGp7Sou1KD83cMnL1ZKNa1UW1VogdhlmeiAACQUhqj/AMI3tsmzlkhUIEgM11FxzxV9t/IFlK9et53TU2cdEXl+ouXn9PhVXlGtveamKyktVVF6mveWl2ltequLyUu0N3i8qL1VRRVmN/XxNNAIrNDpljzO6j8KcMpRsGEoyLSVUVSrR5w9P/bUwte5UhR+1aaVSm12+amFF9ZEVPrtdXrstWFZrLQtJHkM6/Caoa1r1jZqovu5EKJBwVwsgQoFE9Sme6t3H5ZaLtSiAer349BN1ykKjVa9JTT/U3TloMVlovbaCggJt3LhRktS1a1dlZWXFuomYsSxLI0aM0Lx58/TYY4/poYcekiR98MEHuvDCCzVw4EAtXbq0xpvitm3b1LVr1/DtNm3aNKrt0Hy/9c0HDAAAAADRaM5zDE9JkWY/crsk6aTJT8sVgxNky7K0aVeJZq1Yr2+XL9OyTWtVVLldllEkGQ1fpLt1SqaO7dFPJ/Y9WsN79FO/Dl3jYt5zT0mRZj98m4o25ER1XHrPfjF7DQ8HPr9fOQWbgtNg5Wphfq7yohwFYlhJ+8IPK1OGlRIeBRKJ22nX0V1aa1C31hrUra0GdWujbm1T9exXS/WvzxZJkhL9lbpuR2Aqs5fa/UoV9oTw8XecfYxuOfPI+5KmZVkqq6oMhCfBsGRvWWnN+8HwpCh8OxCsRBda4WC5nc5aIyD2jYiINKVT5FET+wk4gvu4nU6CCgAx1yShSEuzaNEirVq1ShdddJFc1YZRFxcX6+6779ZLL72k5ORkrV27Nhzg+Hw+DRgwQDk5OXriiSf04IMPSpI8Ho8uuugiffrpp7rxxhv1n//8p9H9IhQBAAAAEEvNfY4R+sZgU36YXlBYprk5W/TN0hVakLdGO0u3yTQKJaPhC6m7HW4N6tJLJ/Tpr2N7HqWh3XorNSGpyfrcnEJhVdH6NQ3aPx4Ckd2lxVqUl6tF+TnhUSDlnob//Miyy7DSZbMyq40CqX9kjN1mqE+HDA3qHgg/BnVtrd4dM+W01w1NvH5Ttz3/nb5fHpiyO9Ef+KC/eiAyemBn/fvGcRGPj2cen09FFaUqrjYqJTAypVqYUlE9TNl32zxCPhpz2R11pnFKiDhqota0UHXWrAhOE+XaN+VTjamkHE7Z4iBYBnDkiotQ5KOPPtL555+vzMxMHXvssWrbtq0KCgq0aNEiFRYWKiUlRe+++64mTpxY47gFCxZozJgxKi0t1fDhw9W7d2/NnTtX+fn5GjhwoGbPnq309Mb/o9jcJywAAAAAjizxeI6xp6RS89cWaNaKXM3JXaXNhZtkGoWyjBKpgV8uNmSoR9tOGtWnv4b36Kdje/RT51Ztj5hvJzc0GDkSAxGf36812zYGRoEEp8LK21UQVR2GlVxtFEhoLZD6fza6tUvToK5tNLh7Gw3s1kb9O7dSoqvhUzZ5/aZenL5Mb/6wWruK960z0yYtUb855Shdf/ogApEYMk1TpVWVKq7YNyqleniyNxieFIVuV5RqW+FuFVXUP61fdU67PXI4UX09ihr33fumeKq2T901LGpOC5XgdMXFCDgAiIWYhiJbtmzRJ598opycHBUXFytS1YZh6KWXXopVkw2yfv16PfXUU/rll1+Ul5enPXv2yOl0qnv37ho/frzuuOMOde/ePeKxa9eu1aOPPqpvv/1WhYWF6ty5sy666CL9/ve/V0pK3bkjoxGPJywAAAAAmg7nGFJJhUeL1u/QT6vy9cPqlVq3I18+o1CWUSgZDV+rIDMpTcf3PkrH9uin4T36aUCnHnK1kEW/G+NAwciREojsLi0OToMVCEGWblzXiFEgGbXWAql/FEi79KRA+NG1jQZ1a6OB3Vor/QCLnzeU129q4brtKir3KD3JpWG92hOGtAB+09SZf7lPuds373e/Xu066vN7/yz3Yb6+DgAciWIWivz973/XQw89JJ9v33JQoapD366xLEuGYcjvb5pFsw43nLAAAAAAiCXOMeqq8Pi0ZMNO/Zy7VT+sWqOVW9bLY+6WaSuUjIoDVxDksjs1uGuvcEgyrHsftUpJa8Kex159wcjhGoj4/H6t3pqvhfm5WpiXo8X5ucrftT2qOqIZBZKW6NLAboHwY3C3wCiQ9hlH5rRrqN9ni+bo9tefatC+T195p84+ZlQT9wgAEK2YhCKfffaZzjnnHHXp0kW///3v9d///lfffPONvvrqK+Xm5uqtt97SnDlzdN9992nixIkaPXp0LPp+2OOEBQAAAEAscY5xYB6fX8s37tb8tQX6adUGLczPUYVvd3DKrSLJaPgpcvc2HXRcz34a1qOvju3RTz3bdmzx8+zXDkYOp0BkV0lRtcXQc7Rs03pVNNEokASnXf27tA6HH6GF0I+UKdXQOA0dJRLSp31nfXX/X1v8+wIAxJuYhCKnn366Zs2apVWrVqlXr1665ppr9Prrr9cYEfLHP/5RkydP1uzZs3X88ccfbJNHBE5YAAAAAMQS5xjR8/lNrdlSqPlrC/RzzhbNXZuj4qodsmyFwQXcvQ2uKy0xWcf26Kdh3ftqeI++GtK1txJdsZlKKZZCwYikFhuIeP0+rd66MTwV1sK8XG3asyOqOvaNAskMjgJJiTgKxG4z1LdjZo1RIL07ZMjBVFWo5cslP+vON/5Vpzz0wVqkyOzJK/5HZw4Z0aT9AgBEJyahSJs2bTRgwAB9//33kqRrr71Wr732Wo1QxLIs9enTRwMGDNDHH398sE0eEThhAQAAABBLnGMcPNO0tH57keblFuiX3ALNzV2nXWXbZBp7ZRmFsmylDa7LYbPr6E7dgtNtBRZwz8po1YS9bzhPSZEktZhAZGfJ3sAIkLwcLcwLjAKp9HoaXoHlkGGly2ZlymZmyLDS6x0FcrALoQMAgMNbTP7ql5aWqlOnTuH7bnfgmzAlJSVKTU2VFFhX5LjjjtOMGTNi0SQAAAAAADFnsxnq3SFDvTtk6NJTjpJljdamXSWat3a75q/drp9zN2nTnk2ybHurTbkVed1Mn+nX0k3rtXTTer3yw5eSpI6ZbcKjSY7t0U/9OnSVw24/lA9RUvOGIV6/T6u25AdGgQTXA9m8Z2dUdQRGgWRWWwsk8iiQplwIHQAAHJ5iEoq0b99ee/bsqXFfktauXatjjjkmXL53715VVDR8ITsAAAAAAJqTYRjq2jZNXdum6cJRfSRJBYVlmr9uu+bnbte83G3K3bEpGJDsDS7gXllvfVsLd+mTwl36ZOFPkqRkd4KGdu2tYT36aXj3vjqmex+lJSYfksd2qOwoKgyHHwvzcrVs8zpVeRs+LVlgFEhgDZDAKJAMGXLW2Y2F0AEAQEPEJBTp3bu3NmzYEL4/cuRIWZal5557Tv/5z38kSatWrdLMmTPVv3//WDQJAAAAAECzyMpM1tnH9tTZx/aUJO0pqdSCdYGRJPPWbtfKzVvkN/YEQhKjUJZRUu8C7mVVlfopd7l+yl0uKRDC9M3qrOHd9y3g3rV1+8NmgW+Pz6eVW/K0KD83vBbIlsIoRoFYCoz6qLYguqG6o0ASnHYd3aV1cPRHIATpykLoAACgAWISikyYMEEPPvigli9froEDB+q0005Tnz599OKLL2rRokXq0qWLZsyYIa/Xq5tuuikWTQIAAAAA0CK0Sk3Q+KHdNH5oN0lSSYVHi9bvCE+5tSy/QB4zsHB7aNotGb6IdVmWpTXbNmnNtk16K/tbSVLrlHQN79FXw4OjSQZ26Sm3o+5IieawvWiPFublhkOQZZvXN24UiJmxbyqsWqNAqi+EPjg4EoSF0AEAQGPFJBS5/PLLlZycrNCa7Q6HQx9//LEuuugizZ8/X/Pnz5fNZtMtt9yim2++ORZNAgAAAADQIqUmunTKgM46ZUBnSVKFx6cleTs1L7dA89du16INO1TpLdo3ksRWKMsor7e+3aVFmr5snqYvmydJctkdGtSlZyAkCa5P0iaKNUL8pqk353wjSbr8hPGy2RoWLnh8Pq3YskGL83K1MD8wCmRr4a4Gt7tvFEimbFa6DDNThpLrjALp1i4tHH6wEDoAAIg1wwolGU1kzZo12rNnj/r06aM2bdo0ZVOHnVGjRkmSsrOzm7knAAAAAI4EnGMcHjw+v5Zv3K35aws0P3e7FqzfodLKUpnGXlm2wsC1USQZZoPr7N4mKxyQDO/RV33ad6437Pho/mzd9ebTkqQ7xl+l3044Q84Ioy4K9u4JrAOSn6PFeblatnmDPL5oRoE4w1NgBUaBpNcZBdI+IykQfnRto8Hd22hAVxZCBwAATavJQxHUjxMWAAAAALHEOcbhyW+aWr25MBCSBNclKSwrl2UUVwtKCiXD0+A6UxOSwgHJ8B79NKRrb7kcLv3n6yV66ttn5LNKJEmGmaIOieN18cl9NKp/ipZuXBueCmvr3t0NfxCWZFiptdYCqTkKJLQQ+uDubTSoKwuhAwCA5kEo0ow4YQEAAAAQS5xjHBlM09L67UWav7ZA89Zu17zc7dpeVCapXKYttHj73uAC7g2r026zKcnZSqUVkmWrNeWVmSQZFfUuBh/RAUaBsBA6AABoqZiUEwAAAACAFsRmM9S7Q4Z6d8jQJScfJcuytHl3qX4Jrkkyf+12bdpVIkvewLok1YISGf6IdfpNUyVVu6RIM2rZ6l/PRFKEUSCZMpQUHgVitxnq24mF0AEAwOGBUAQAAAAAgBbMMAx1aZOqLm1SdeGoPpKkgsIyzV+3XfNzAyHJ2oK9smTKMkrCC7ibtr2BESDRqjEKJDM4CmTfxwcshA4AAA5n/NcCAAAAAMBhJiszWWcf21NnH9tTkrSnpFIL1m0Pr0myevMemX5LliqDAUmhTNum+hdvtxyy+/vLVmsUSGgh9NCFhdABAMDhjlAEAAAAAIDDXKvUBI0f2k3jh3aTJJVUeLRo/Q7NW7td3y7J19qda2Qa+fVXYPhkyC6bknX60G469/heLIQOAACOSIQiAAAAAAAcYVITXTplQGedMqCzTuiXpd88+/kBj/Hb1spmZuk3o4/SiL4dDkEvAQAADj1WPQMAAAAA4Ai2vXSDLFvpAfezbKVKTt6jYb3aH4JeAQAANA9CEQAAAAAAjlB+09SUbz9s8P4213rZjSbsEAAAQDNr1PRZ48aNa3SDhmFoxowZjT4eAAAAAAA0zPRl85S/q0BOu0M+05RlWvXua9gMFVfu1dfL5unMISMOYS8BAAAOnUaFIrNmzYpYbhiGLCvyP1ihbYbBV04AAAAAADgUzhwyQmcOeVOS5PWbenH6Mr35w2rtKq4I79MmLVG/OeUoXX/6IDntTCgBAACObI0KRTZs2FCn7J///Kf+/e9/68ILL9Tll1+u7t27S5Ly8vL05ptv6oMPPtBtt92mO++882D6CwAAAAAAGsFpt+mWM4fo+tMHaeG67Soq9yg9yaVhvdoThgAAgLjRqFCkW7duNe6/++67evrpp/XBBx/o3HPPrbFt8ODBOuecc/TJJ5/oggsu0IgRI+ocDwAAAAAADg2n3aYRfTs0dzcAAACahWHVN99VFEaMGCG3260ffvhhv/udcsopqqys1C+//HKwTR4RRo0aJUnKzs5u5p4AAAAAOBJwjgEAAADsX0zGx65YsUJdunQ54H5dunTRypUrY9EkAAAAAAAAAABAVGISirjdbi1cuPCA+y1cuFButzsWTQIAAAAAAAAAAEQlJqHIuHHjlJOTo3vvvVder7fOdq/Xq/vvv185OTkaN25cLJoEAAAAAAAAAACISkzWFMnNzdXxxx+v4uJiZWVl6fzzz1e3bt1kGIby8/P14Ycfatu2bUpLS9PPP/+svn37xqLvhz3m+wUAAAAQS5xjAAAAAPvniEUlffr00axZs3TllVdq2bJlmjJligzDkCSFMpeBAwfq9ddfJxABAAAAAAAAAADNIiahiCQNGTJES5Ys0Q8//KAffvhBW7ZskSR16tRJJ598skaPHh2rpgAAAAAAAAAAAKIWs1Ak5JRTTtEpp5wS62oBAAAAAAAAAAAOSkwWWq+tqqpK27Zt0549e5qiegAAAAAAAAAAgKjFNBR54YUXNGzYMKWkpKhz58669957w9vef/99XXDBBVq7dm0smwQAAAAAAAAAAGiQmIQipmlq0qRJuvnmm7Vy5UodddRR4QXWQwYPHqyPPvpI06ZNi0WTAAAAAAAAAAAAUYlJKPLss8/q/fff1+mnn668vDwtW7aszj59+vRRz5499eWXX8aiSQAAAAAAAAAAgKjEJBR55ZVX1LZtW02bNk1ZWVn17jdw4EDl5+fHokkAAAAAAAAAAICoxCQUWb16tUaOHKnU1NT97peSkqIdO3bEokkAAAAAAAAAAICoxCQUsdls8nq9B9xv06ZNSk5OjkWTAAAAAAAAAAAAUYlJKNKvXz8tXLhQFRUV9e5TWFioxYsXa/DgwbFoEgAAAAAAAAAAICoxCUV+/etfa8eOHbrzzjtlmmbEfe69916Vlpbq0ksvjUWTAAAAAAAAAAAAUXHEopLbb79db7/9tl588UXNnTtXZ599tqTAWiN/+tOf9PHHH2vhwoUaMWKErr322lg0CQAAAAAAAAAAEJWYhCIJCQn65ptvdM011+izzz7TsmXLJElz587V3LlzJUkTJ07UG2+8IYcjJk0CAAAAAAAAAABEJWYJRevWrfXJJ59o6dKlmj59ujZs2CDTNNWlSxedfvrpOvbYY2PVFAAAAAAAAAAAQNRiPmxj8ODBLKYOAAAAAAAAAABanJgstA4AAAAAAAAAANDSxSQUeeutt9SzZ099/fXX9e7z9ddfq2fPnpo2bVosmgQAAAAAAAAAAIhKTEKRqVOnqqSkROPGjat3n7Fjx6q4uFhvvPFGLJoEAAAAAAAAAACISkxCkWXLlmnw4MFyOp317uNyuTRkyBAtXbo0Fk0CAAAAAAAAAABEJSahyM6dO9WhQ4cD7peVlaUdO3bEokkAAAAAAAAAAICoxCQUSU9P18aNGw+43+bNm5WcnByLJgEAAAAAAAAAAKISk1Bk+PDhmjt3rlavXl3vPqtXr9acOXM0fPjwWDQJAAAAAAAAAAAQlZiEIjfccIN8Pp9+9atfad68eXW2z58/X7/61a9kmqauv/76WDQJAAAAAAAAAAAQFUcsKjn//PP1m9/8Rm+++aZGjhyp/v37q2/fvpKknJwcrVq1SpZl6dJLL9WkSZNi0SQAAAAAAAAAAEBUYhKKSNLrr7+uAQMG6K9//atWrlyplStXhrdlZmbqvvvu0wMPPBCr5gAAAAAAAAAAAKJiWJZlxbJCr9erBQsWhBde79q1q4YPHy6n0xnLZo4Io0aNkiRlZ2c3c08AAAAAHAk4xwAAAAD2L2YjRUKcTqdGjhypkSNHxrpqAAAAAAAAAACARovJQusAAAAAAAAAAAAtXUxHivz444+aNWuWtm7dqqqqqoj7GIahl156KZbNAgAAAAAAAAAAHFBMQpHS0lJdcMEFmjFjhg60RAmhCAAAAAAAAAAAaA4xCUUeeeQRffvtt+revbtuvfVWHXXUUUpNTY1F1QAAAAAAAAAAADERk1Dk/fffV+vWrTV37ly1a9cuFlUCAAAAAAAAAADEVEwWWt+5c6dOOukkAhEAAAAAAAAAANBixSQU6dKlywHXEgEAAAAAAAAAAGhOMQlFLrvsMn3//ffau3dvLKoDAAAAAAAAAACIuZiEIg8//LCGDh2qiRMnatWqVbGoEgAAAAAAAAAAIKZistD6GWecIa/Xq7lz52rQoEHq2rWrunbtKputbuZiGIZmzJgRi2YBAAAAAAAAAAAaLCahyKxZs8K3TdNUXl6e8vLyIu5rGEYsmgQAAAAAAAAAAIhKTEKRDRs2xKIaAAAAAAAAAACAJhOTUKRbt26xqAYAAAAAAAAAAKDJxGShdQAAAAAAAAAAgJaOUAQAAAAAAAAAAMSFmEyfFTJt2jS9//77ysnJUXFxsSzLqrOPYRhat25dLJs9IK/Xq5kzZ+qzzz7TrFmztG7dOvn9fnXu3FkTJkzQ/fffH3EKsAMtCp+dna2RI0c2VbcBAAAAAAAAAEAMxSQU8fv9uuCCC/TZZ59FDEKkQMBQ37am9v3332vChAmSpC5duuj000+XJP3yyy+aMmWKpk6dqq+++kqjRo2qc2xycrIuuuiiiPW2bdu26ToNAAAAAAAAAABiKiahyNNPP61PP/1UZ555pp566in96U9/0tSpU1VZWam1a9dq6tSp+uc//6m77rpLjz32WCyajIrNZtNFF12ku+++u0bwUVlZqVtuuUWvvvqqLr30UuXm5srpdNY4tk2bNnr11VcPcY8BAAAAAAAAAECsxWRNkbfeekvp6el6++231bt3b9lsgWqdTqf69++vxx57TB988IH+3//7f3rrrbdi0WRUxo0bp/fee6/OSJCEhARNmTJF6enpys/P15w5cw553wAAAAAAAAAAwKERk1Bk9erVGjFihNLS0iTtW4vD7/eH9znjjDN03HHH6d///ncsmoyZxMRE9e3bV5K0devWZu4NAAAAAAAAAABoKjFbU6RVq1bh+0lJSZKkvXv3qnXr1uHyXr166bPPPotFkzHj9/uVl5cnScrKyqqzvaysTI8//rjy8/OVkJCgAQMG6Nxzz1X79u0PcU8BAAAAAAAAAMDBiEko0rFjR23bti18v1u3bpKkpUuXauzYseHyDRs2yOGISZMx88Ybb2jnzp1q27atTjjhhDrbd+3apYcffrhG2R133KHHH39cd911V4PaiLSAuyQtX75cHTp00EcffaTTTjstXP7jjz+qrKxMY8aMUUJCgiRp4cKF2rFjh4477rhw0LRmzRpt2LBBAwYMUJcuXSRJmzZt0ooVK9S9e3cdddRRkqTCwkL9/PPPatu2rYYPHy5Jqqqq0syZM5WUlKRTTjkl3PaMGTPk9Xp1+umnh6dBmzt3rvbu3asTTzxRqampkqRly5Zpy5YtOuaYY8IB0fr165WTk6O+ffuqZ8+ekqTt27dr0aJF6tSpkwYNGiRJKikp0U8//aSMjAyNHDlSkmSapqZPny6n06lTTz013J8ffvhB5eXlGjt2rNxutyRpwYIF2rlzp0aMGKHMzExJgdFKeXl5EZ+LHj16qF+/fpKk3bt3a968eWrXrp2GDRsmKbC2zKxZs5ScnKyTTz453Pa3334rn8+nM844I1yWnZ2toqIinXzyyUpOTpYU+DnfunWrhg0bpnbt2kmS1q1bp9zcXPXr1089evSQJBUUFGjx4sXq3LmzBg4cWOO5yMzM1IgRIyQFgrpvvvlGLpdL48aNC7f9/fffq6KiQuPGjZPL5ZIkzZ8/X7t27arxXKxatUr5+fkaNGiQOnXqJEnKz8/XqlWr1LNnz/DIqF27dmn+/Plq3769jjnmGElSRUWFvv/+e6WkpOikk04Kt/3NN9/INE1NmDAhXDZnzhwVFxfXeC6WLFmibdu2afjw4Wrbtq0kae3atVq7dq2OOuoode/eXZK0bds2LVmyRF26dNGAAQMkSUVFRcrOzlarVq10/PHHS5J8Pp++/fZbud3uGu8ls2bNUmVlpU499dTwOkDz5s3T7t27NWrUKKWnp0uSVq5cqY0bN2rw4MHq2LFjjeeiV69e6tOnjyRp586dWrBggbKysjR06FBJUnl5uX744QelpqbqxBNPDLc9ffp0SdLpp58eLvvpp59UUlKiU045JRwKL168WAUFBTWei9zcXK1bt079+/cPv09u3bpVS5cuVdeuXXX00UfXeC5at26t4447TpLk9Xo1Y8YMJSQkaMyYMeG2Z86cqaqqKp122mnh99dffvlFe/bsqfFcrFixQps2bdKQIUPUoUMHSVJeXp5Wr16t3r17q3fv3jWeiw4dOmjIkCGSAuHwjz/+qLS0tBrvk19//bVsNpvGjx8fLps9e7ZKS0s1evRoJSYmSpIWLVqk7du369hjj1WbNm0kSTk5OVq/fn2N52LLli1atmyZunXrpv79+0va9/7Vpk0bHXvssZIkj8ej7777TomJiRo9enS47e+++04ej0fjx4+X3W6XJP38888qLCys8f61fPlybd68WUOHDg2H4Rs2bNCaNWvUp08f9erVS5K0Y8cOLVy4UB07dtTgwYNrPBfp6ek13tu/+uorORwO3st5L+e9nPdy3sv3815+0UUXCbHDOQZ/lyT+LvF3ib9LnGPwXs57Oe/lvJfH93t5Y88xYjJ91uDBg7VmzZrw/bFjx8qyLD366KMqKSmRJL355puaO3du+I2uJcjLy9M999wjSXrsscfCb6IhV1xxhb744gtt3rxZZWVlWrp0qW677TZ5vV7dfffd+s9//tMc3QYAAAAAAAAAAI1gWJZlHWwlL730km644QZ988034dT27LPP1hdffCGHw6HU1FTt3btXUiBNrZ7gNZdQYrl06VJNmjRJ06ZNa/Cxzz33nG655Ra1atVKW7durROmNFToGwDZ2dmNOh4AAAAAquMcAwAAANi/mIwU+c1vfqNly5aFh85I0rvvvqtbb71VrVu3VmlpqQYOHKh33nmnRQQilZWVOuecc7R06VKdeuqpeuONN6I6/sYbb1Tbtm21Z88ezZ07t4l6CQAAAAAAAAAAYikmC3yEFiCvLjk5Wf/+97/173//OxZNxIzX69WFF16o77//XiNHjtTHH38c9UgPm82m3r17a+fOndq6dWsT9RQAAAAAAAAAAMRSTEaKHC5M09Tll1+uL774QkOGDNEXX3wRXvAnWoWFhZLU6OMBAAAAAAAAAMChFTehiGVZuuGGGzRt2jT169dP06dPV2ZmZqPqWr58uVavXi1JGj58eCy7CQAAAAAAAAAAmkijps8aN25coxs0DEMzZsxo9PGNdc899+jll19Wjx49NGPGDLVr126/+7/22msaNGiQhg0bVqN84cKFuuyyyyRJF154oTp16tRkfQYAAAAAAAAAALHTqFBk1qxZjW7QMIxGH9tYH3/8sf75z39Kkrp3766HH3444n7nnXeezjvvPEnShx9+qKuvvlr9+vXT0UcfLZfLpbVr12rRokUyTVPDhw/X888/f6geAgAAAAAAAAAAOEiNCkU2bNgQ6340qdD6H5I0c+bMevfr3r17OBS58sorlZSUpEWLFmnWrFkqKSlRWlqaTjnlFP3617/WddddJ5fL1dRdBwAAAAAAAAAAMWJYlmU1dyfi1ahRoyRJ2dnZzdwTAAAAAEcCzjEAAACA/YubhdYBAAAAAAAAAEB8IxQBAAAAAAAAAABxoVFritRn2rRpev/995WTk6Pi4mJFmpnLMAytW7culs0CAAAAAAAAAAAcUExCEb/frwsuuECfffZZxCBECoQhLF8CAAAAAAAAAACaS0ymz3r66af16aef6owzzlBOTo6uuOIKGYahqqoqrVixQr/73e/kdrv1u9/9TqZpxqJJAAAAAAAAAACAqMRkpMhbb72l9PR0vf3220pLS5PNFshanE6n+vfvr8cee0wnn3yyzjrrLA0YMECXXXZZLJoFAAAAAAAAAABosJiMFFm9erVGjBihtLQ0SYGpsqTAtFohZ5xxho477jj9+9//jkWTAAAAAAAAAAAAUYlJKOL3+9WqVavw/aSkJEnS3r17a+zXq1cvrVixIhZNAgAAAAAAAAAARCUmoUjHjh21bdu28P1u3bpJkpYuXVpjvw0bNsjhiMmMXQAAAAAAAAAAAFGJSSgyePBgrVmzJnx/7NixsixLjz76qEpKSiRJb775pubOnatBgwbFokkAAAAAAAAAAICoxCQUmThxogoKCjRjxgxJ0rHHHquJEyfqxx9/VOvWrdW6dWtdeeWVMgxDf/jDH2LRJAAAAAAAAAAAQFRiEor85je/0bJlyzRkyJBw2bvvvqtbb71VrVu3VmlpqQYOHKh33nlHY8eOjUWTAAAAAAAAAAAAUTEsy7KauxPxatSoUZKk7OzsZu4JAAAAgCMB5xgAAADA/sVkpAgAAAAAAAAAAEBL52jKyouLi7V27VolJyerX79+TdkUAAAAAAAAAADAfh1UKLJkyRJ9//338vl86t+/v8444wwZhiGv16u77rpLzz//vPx+vySpT58+eu211zRixIiYdBwAAAAAAAAAACAajQ5Fbr/9dk2ZMqVG2bBhw/Tdd9/poYceqrMtJydHZ5xxhpYvX65OnTo1tlkAAAAAAAAAAIBGadSaIu+8846eeeYZSdJJJ52kCy64QD169NDChQv18MMP6/nnn9cZZ5yhZcuWqaysTEuXLtXpp5+uoqIi/eMf/4jpAwAAAAAAAAAAAGiIRo0UeeGFF2QYhj799FNNnDhRkuT3+3XVVVfpmWeeUUZGhqZNm6aUlBRJ0sCBA/Xee++pe/fumj59eux6DwAAAAAAAAAA0ECNGimyZMkSHXfcceFARJLsdrseeeQRWZal4447LhyIhKSmpuq4445TXl7eQXUYAAAAAAAAAACgMRoVihQWFqpXr151ykNlHTp0iHhcVlaWysvLG9MkAAAAAAAAAADAQWlUKGJZltxud51yp9MpSTIMI+Jx9ZUDAAAAAAAAAAA0tUaFIgAAAAAAAAAAAIebRi20LkkFBQX64YcfotpWUFDQ2OYAAAAAAAAAAAAOimFZlhXtQTab7aCmwvL7/Y0+9kgyatQoSVJ2dnYz9wQAAADAkYBzDAAAAGD/GjVSpGvXrqwPAgAAAAAAAAAADiuNCkXy8vJi3A0AAAAAAAAAAICmxULrAAAAAAAAAAAgLhCKAAAAAAAAAACAuEAoAgAAAAAAAAAA4gKhCAAAAAAAAAAAiAuEIgAAAAAAAAAAIC4QigAAAAAAAAAAgLhAKAIAAAAAAAAAAOICoQgAAAAAAAAAAIgLhCIAAAAAAAAAACAuEIoAAAAAAAAAAIC4QCgCAAAAAAAAAADiAqEIAAAAAAAAAACIC4QiAAAAAAAAAAAgLhCKAAAAAAAAAACAuEAoAgAAAAAAAAAA4gKhCAAAAAAAAAAAiAuEIgAAAAAAAAAAIC4QigAAAAAAAAAAgLhAKAIAAAAAAAAAAOICoQgAAAAAAAAAAIgLhCIAAAAAAAAAACAuEIoAAAAAAAAAAIC4QCgCAAAAAAAAAADiAqEIAAAAAAAAAACIC4QiAAAAAAAAAAAgLhCKAAAAAAAAAACAuEAoAgAAAAAAAAAA4gKhCAAAAAAAAAAAiAuEIgAAAAAAAAAAIC4QigAAAAAAAAAAgLhAKAIAAAAAAAAAAOICoQgAAAAAAAAAAIgLhCIAAAAAAAAAACAuEIoAAAAAAAAAAIC4QCgCAAAAAAAAAADiAqEIAAAAAAAAAACIC4QiAAAAAAAAAAAgLhCKAAAAAAAAAACAuEAoAgAAAAAAAAAA4gKhCAAAAAAAAAAAiAuEIgAAAAAAAAAAIC4QigAAAAAAAAAAgLhAKAIAAAAAAAAAAOICocgBeDwePfHEExowYIASExPVtm1bXXDBBVq4cGFzdw0AAAAAAAAAAESBUGQ/PB6PJkyYoIceeki7du3Sr371Kx111FH68MMPNXLkSH399dfN3UUAAAAAAAAAANBAhCL78ec//1mzZs3Scccdp9zcXE2bNk0//vij3nrrLXm9Xl1++eUqKSlp7m4CAAAAAAAAAIAGIBSph8/n05NPPilJmjJlitLS0sLbLr30Uk2cOFG7du3Syy+/3Ew9BAAAAAAAAAAA0SAUqcdPP/2kPXv2qEePHjr22GPrbL/44oslSR9//PGh7hoAAAAAAAAAAGgEQpF6LF68WJI0fPjwiNuHDRsmSVqyZMmh6hIAAAAAAAAAADgIjubuQEuVn58vSercuXPE7aHyPXv2qLS0VCkpKfXWNWrUqIjl8+fPl9vt1vHHH3+QvQUAAABatiFDhuiFF15o7m4cMTjHAAAAQLxr7DkGoUg9SktLJUnJyckRt1cPQUpKSvYbitTH4XAoOTlZdru9cZ08SMuXL5ckDRw4sFnaR8PwOh0+eK0OD7xOAOIR733xg3MMNBSv1eGB1+nwwOsEIB4dzu99hCKHQHZ2dnN3IaLQt8taav8QwOt0+OC1OjzwOgGIR7z3HXla6mvJz9rhg9fq8MDrdHjgdQIQjw7n9z7WFKlHaORHWVlZxO2hkSSSlJqaekj6BAAAAAAAAAAAGo9QpB7dunWTJG3evDni9lB5q1atGjV1FgAAAAAAAAAAOLQIReoxdOhQSdKCBQsibl+4cKGkwGIuAAAAAAAAAACg5SMUqceJJ56oVq1aacOGDZo/f36d7e+++64k6dxzzz3UXQMAAAAAAAAAAI1AKFIPh8OhO++8U5J06623qri4OLzt7bff1hdffKE2bdro2muvbaYeAgAAAAAAAACAaBiWZVnN3YmWyuPxaMKECZo1a5batWun0aNHq6CgQD/++KOcTqc++eQTnXHGGc3dTQAAAAAAAAAA0ACEIgfg8Xj0t7/9TVOnTtX69euVkpKik046Sf/7v/+rYcOGNXf3AAAAAAAAAABAAxGKAAAAAAAAAACAuMCaIgAAAAAAAAAAIC4QigAAAAAAAAAAgLhAKAIAAAAAAAAAAOICoQgAAAAAAAAAAIgLhCIAAAAAAAAAACAuEIoAAAAAAAAAAIC4QCgCAAAAAAAAAADiAqEIAAAAAAAAAACIC4QiAAAAAAAAAAAgLhCKAAAAAAAAAACAuEAoAgAAAAAAAAAA4gKhCAAAAAAAAAAAiAuEIgAAAAAAAAAAIC4QigAAAAAAAAAAgLhAKAIAAAAAAAAAAOICoQgAAAAAAAAAAIgLhCIAAAAAAAAAACAuEIoAAAAAAAAAAIC4QCgCAAAAAAAAAADiAqEIAAAAAAAAAACIC4QiAAAAAAAAAAAgLhCKAAAAAAAAAACAuEAoAgAAAAAAAAAA4gKhCAAAAAAAAAAAiAuEIgAAAAAAAAAAIC4QigAAAAAAAAAAgLhAKAIAAAAAAAAAAOICoQgAAAAAAAAAAIgLhCIAAAAAAAAAACAuEIoAAAAAAAAAAIC4QCgCAAAAAAAAAADiAqEIAAAAAAAAAACIC4QiAAAAAAAAAAAgLhCKAAAAAAAAAACAuEAoAgAAAAAAAAAA4gKhCAAAAAAAAAAAiAuEIgAAAAAAAAAAIC4QigAAAAAAAAAAgLhAKAIAAAAAAAAAAOICoQgAAAAAAAAAAIgLhCIAAAAAAAAAACAuEIoAAAAAAAAAAIC4QCgCAAAAAAAAAADiAqEIAAAAAAAAAACIC4QiAAAAAAAAAAAgLhCKAAAAAAAAAACAuEAoAgAAAAAAAAAA4gKhCACgRejevbsMw9Crr74aszoNw5BhGJo1a1bM6gQAAABweOAcAwAQCaEIABzmHn300fA/5tUvbrdbHTt21JlnnqlXX31Vfr+/Wfr35JNP6tFHH9XixYubpf0jUaye0+nTp+vBBx/U+PHj1bt3b2VkZMjlcql9+/YaN26cnn76aVVUVMSm0wAAADhscI4Rf5r6OR0/fnz452jMmDFN0gYANJSjuTsAAIid9u3bh2+XlJRo27Zt2rZtm7766is9//zz+vLLL5Wenn5I+/Tkk08qPz9f3bt319ChQ+vdr1evXkpISIhp//r16ydJSkpKilmdLUFDn9MD+eMf/6iffvopfD85OVkul0s7duzQjh07NHPmTD311FP68ssv1adPnxj0HAAAAIcbzjFq4hwjei+88IK+/fbbmNYJAAeDkSIAcAQpKCgIX0pLS7V27Vr9+te/liRlZ2fr5ptvbuYe1m/GjBlavXq1zj///JjVuXr1aq1evVrHH398zOo8kpx99tmaMmWKFi9erNLS0vBl586d+te//qXExEStW7dO559/vkzTbO7uAgAAoBlwjlET5xjR2bx5s+69915lZGSof//+zd0dAJBEKAIARyzDMNSrVy+98847OvHEEyVJ7777rrZv397MPUNL8eCDD+qWW27RkCFDlJycHC5v06aNbr/9dj355JOSpBUrVig7O7uZegkAAICWgnMMROumm25ScXGx/va3v6ldu3bN3R0AkEQoAgBHPMMwdOWVV0qSLMvS/PnzJUnl5eV65513dNVVV2no0KFq166dXC6XsrKyNHHiRL333nv11jlr1qzwfLCStHjxYl1xxRXq0qWLXC6Xhg4dGp6HOD8/X5J0zTXX1JmTOC8vL1xnQxZBXLRokW644Qb169dPKSkpSk1N1VFHHaVLL71U77//fp3RDPUtgpiXl1ejD+vXr9f111+vrl27KiEhQZ06ddJVV12ltWvX1tuX5cuX6//+7/906qmnqlevXkpKSlJqaqoGDRqke+65R5s3b6732DFjxsgwDD366KOyLEsvvPCCRowYobS0NKWmpmrUqFGaOnVqneOifU4P1siRI8O39/d4AAAAEF84x+AcoyFee+01ffHFFxo7dqyuu+66RtUBAE2BNUUAIA507tw5fLuoqEiSNG3aNF1zzTWSAv/Yp6WlyeVyafv27fryyy/15Zdf6tJLL9Wbb74ZPjGJ5MMPP9Qll1wij8ej1NRUORyBPy0pKSlq3769du7cKdM0lZaWpsTExBrH2u32BvXfNE09+OCD+utf/xouS0xMVEJCgnJzc7VmzRq98847KiwsVEZGRoPqDJk3b56uv/56FRcXKzk5WXa7XVu3btXrr7+u9957T5988olOO+20OsedffbZ4RMHt9ut5ORk7d27V8uXL9fy5cv1yiuv6PPPP9eoUaPqbdvv9+v888/Xxx9/LIfDoaSkJJWUlGju3LmaO3eucnNz9X//93/h/WP5nDbEjz/+GL7dq1evmNULAACAwx/nGPXjHEPatm2b7rrrLiUmJur555+P+ngAaEqMFAGAOLBhw4bw7VatWkmSMjMzdffdd+uHH35QSUmJ9u7dG15P4h//+IdSU1P19ttv65lnntlv3VdeeaXGjh2rFStWqLi4WOXl5Zo2bZruvfdeFRQUqEuXLpKkp556qsZ8xNW3Hcgf/vCH8MnKZZddpuXLl6u8vFx79uxRcXGxvvzyS/3617+WzRb9n7Ubb7xRXbp00U8//aTS0lKVlZXpxx9/1FFHHaWKigpNmjRJW7durXPcKaecopdfflkbNmxQRUWFdu/ercrKSs2ZM0enn366CgsLNWnSJFVWVtbb9jPPPKPvvvtOL7/8soqLi1VUVKRNmzbpV7/6lSRp8uTJys3NDe8fy+e0PhUVFcrJydHjjz+ue+65J/xYjz322IOqFwAAAEcWzjHqxzmGdMstt6iwsFCPPvqoevfuHfXxANCkLADAYe0Pf/iDJcmq7y3d6/VagwYNsiRZdrvd2r17d4PqnTZtmiXJ6t27d51tM2fODLc5fPhwy+v11ltPt27dLEnWK6+8st/26tsvNzfXstvtliTrjjvuaFDfQ0J9nDlzZo3yDRs2hLdlZmZa27Ztq3Psli1brPT0dEuSddttt0XVrs/nCz/nU6dOrbN99OjR4fa/+eabOtsrKyutjh07WpKsyZMn19ne0Oe0oXJzc8P9qX4xDMM677zzrD179sSkHQAAABweOMeoH+cYB/bWW29ZkqxjjjmmxusY6uPo0aMPug0AOBiMFAGAI1RxcbF++uknTZgwQcuWLZMkXX311eFvcR3I2WefLUlau3atCgoK6t3vgQceCA9nbwqvvfaa/H6/MjIy9Pjjj8e8/ptvvllZWVl1yjt27Kgbb7xRkvTOO+9EVafdbteZZ54pqeYUVLWdcMIJEYfNu91uTZgwQZK0dOnSqNpuDIfDofbt26t9+/ZKSEgIl19yySX685//rMzMzCbvAwAAAFo+zjEaJp7PMXbu3Kn/+Z//kd1u14svvtikryMANBahCAAcQaovhpeenq6TTjpJ3333nSRpwoQJ+te//lVj/507d+qPf/yjTjzxRLVp00ZOpzN8fFJSUni//S3od9JJJzXNgwmaPXu2JOnUU0+t0adYiXTCEDJu3DhJ0q5duyIuLhiaE7l3795KSUmp8fz/5S9/kbT/527EiBH1buvYsaMkac+ePQ15GAele/fu4aHx5eXl2rhxox566CF9+OGHGjRokF566aUm7wMAAABaJs4xohfP5xi//e1vtWvXLt19990aNmxYk7UDAAeDuBYAjiDt27cP33a5XGrdurUGDx6sSZMm6ayzzqqxmOHcuXM1ceJEFRYWhsuSk5OVmZkZnjd3+/btkqSysrJ622zXrl2sH0YNoW+Qde/evUnq79SpU73bqi8euX379nAfTNPUVVddpalTp4a32+12ZWZmyuVySVJ47uD9PXdpaWn1bgt9o8rr9TboccSKYRjq0qWLHnvsMQ0fPlwXXnihbrrpJh1//PEaNGjQIe0LAAAAmh/nGNGL13OMDz74QO+995569epVYzF3AGhpCEUA4AiyvyHo1fl8Pl166aUqLCzUoEGD9Pjjj+vkk09Wenp6eB+/3x/+p9myrHrrstvtB9fpA6h+ktVSvPzyy5o6dapsNpseeOABXXXVVerdu3eN5+L3v/+9Jk+evN/nrqW74IIL1LVrV23cuFEvvviinnrqqebuEgAAAA4xzjEOjcP9HMPj8ejWW2+VJP3973+X3+9XaWlpjX38fn/4OrQtISGBKbYAHHK86wBAHMrOzlZeXp5sNps+//xzdenSpc4+27Zta4ae1ZWVlaVVq1ZFHFoeC1u2bFG/fv3q3RZS/Rtyb7/9tiTpuuuuq3cO4pby/B2sTp06aePGjcrNzW3urgAAAKAF4xxjn3g8xygvLw+PAjrvvPP2u+/s2bOVmpoqSXrllVd09dVXN3HvAKAm1hQBgDi0adMmSVLbtm0jnqxI0jfffBOTtkLD5Bv7baYTTzxRkjRjxgyVl5fHpE/VheZD3t+21q1b1xhaH3r+6psj1zTN/dZ7sA72OW0oy7K0YcMGSQqftAAAAACRcI6xD+cYANCyEYoAQBzKyMiQJO3YsSPicPiioiJNnjw5Jm2FhstXn1c4GldffbXsdrv27t2rhx56KCZ9qu65557Tjh076pQXFBTo+eeflyRdcsklNbaFnr8lS5ZErHPKlCnhMKEpHOxzKgWmNziQV155JfzzMWbMmEa3BQAAgCMf5xj7xOM5RkZGhizL2u9l9OjRkqTRo0eHyxglAqA5EIoAQBw68cQTlZKSIsuyNGnSJK1atUpS4FtBP/zwg0aPHq2ioqKYtBVanPu///2vdu/eHfXxvXr10gMPPCBJeuqpp3T55Zdr5cqV4e1lZWX65JNPdM4556i4uDjq+v1+v0477TTNnTs3XPbTTz/p1FNP1d69e5WWlqYHH3ywxjFnnHGGJOmFF17Qs88+q6qqKknS7t279eijj+qOO+5Q69ato+5LQx3scypJn3/+ucaPH6+3335bW7duDZdblqWcnBw98MADuummmyRJffr04WQFAAAA+8U5xj7xeo4BAIcLQhEAiEPp6en6xz/+ISkwn+vRRx+t1NRUJScna/To0Vq3bp3ee++9mLR14403ymazKTs7W+3bt1eHDh3UvXt3de/eXZs3b25QHX/605905513SpLefPNNDRgwQMnJyWrdurXS0tJ07rnn6tNPP5VpmlH37/nnn1d+fr5GjRqllJQUpaSk6KSTTtLKlSuVkJCgadOmqXPnzjWOueeee3T00UfL7/fr1ltvVVJSklq1aqW2bdvq//7v/3T22WfrlltuibovDRWL59SyLH377be67LLL1KlTJyUlJalt27ZKSkpSv3799Je//EU+n0/Dhg3Tt99+q8TExCZ7PAAAADj8cY6xT7yeYwDA4YJQBADi1A033KCvvvpKp556qtLS0uT1etW+fXvdcMMNWrRokcaOHRuTdk466SR9+eWXOuOMM5SZmamdO3cqPz9f+fn5DZrCSQrMb/vPf/5Tc+fO1ZVXXqkePXrINE35fD716dNHl112mT788EOlpaVF3b/jjjtOixYt0rXXXqtWrVrJ6/UqKytLV1xxhZYsWaIJEybUOSY9PV1z5szRPffco549e8put8swDJ144ol64YUX9NFHH8lut0fdl4aKxXM6ZswYvfLKK7rqqqs0ePBgpaamau/evbLb7erdu7cuvvhiTZs2TfPmzVPXrl2b7LEAAADgyME5RkC8nmMAwOHCsFhBCQAQZ/Ly8tSjRw9J0oYNG2oscAgAAAAA0eIcAwAOH4wUAQAAAAAAAAAAcYFQBAAAAAAAAAAAxAVCEQAAAAAAAAAAEBcOu1Bk4cKF+stf/qJJkyape/fuMgxDhmEoLy9vv8etW7dOV1xxhTp06KCEhAT17t1bDz74oEpLS+s9xrIsPffccxo+fLiSk5PVqlUrnX766ZoxY0aMHxUAAAAAAAAAAGhqh91C6+edd54+/vjjOuX7W8Rq4cKFGjNmjEpKSjRs2DD17t1bP//8s/Lz8zVw4EDNnj1b6enpNY6xLEtXXHGF3nzzTaWmpmr8+PEqKSnRd999J9M09cILL+i6665riocIAAAAAAAAAACawGE3UmTUqFF65JFH9NFHH2nz5s1q3779fvf3+/269NJLVVJSoieeeEILFizQu+++q5ycHJ199tlavny57r///jrHTZ06VW+++aZ69Oih1atX6/3339f06dP13XffyeFw6NZbb1V+fn5TPUwAAAAAAAAAABBjh91IkdqysrK0ffv2ekeKfPDBB7rwwgs1cOBALV26VIZhhLdt27ZNXbt2lSQVFBSodevW4W2DBw/WsmXL9N577+miiy6qUedvf/tbTZkyRXfeeaf++c9/Ns0DAwAAAAAAAAAAMXXYjRSJ1qeffipJmjRpUo1ARJI6dOigk08+WT6fT59//nm4PC8vT8uWLVNCQoLOOeecOnVefPHFkhRxGi8AAAAAAAAAANAyHfGhyOLFiyVJw4cPj7h92LBhkqQlS5aEy0K3Bw4cKJfLVe8xGzZsUHFxcSy7CwAAAAAAAAAAmoijuTvQ1ELrfnTu3Dni9lB59fVBDnRMSkqK0tPTVVRUpI0bN2rgwIH77cOoUaMilq9fv17nnHOOXnjhhf0/CAAAAACohnMMAAAAoHGO+FCktLRUkpScnBxxe0pKiiSppKSkwceEjisqKqpxXLTKy8u1ZMmScHsAAADAkSr0fzeaFucYAAAAiBeNPcc44kORliA7OztieejbXZwgAgAAAIgG5xgAAABA4xzxa4qETgbKysoibg99gyo1NbXBx9R3HAAAAAAAAAAAaLmO+FCkW7dukqTNmzdH3B4qD+3XkGNKS0tVVFQkSeratWvM+goAAAAAAAAAAJrOER+KDB06VJK0YMGCiNsXLlwoSRoyZEi4LHR7+fLl8ng89R7To0cPpaWlxbK7AAAAAAAAAACgiRzxocivfvUrSdJ7770ny7JqbNu2bZt+/PFHORwOTZw4MVzevXt3DRo0SJWVlfrkk0/q1Pnuu+9Kks4999wm7DkAAAAAAAAAAIilIz4UOeecc9S3b18tX75cf/7zn8PlHo9HN910k3w+n6699lq1adOmxnH33XefJOn+++/X1q1bw+U//PCDXnjhBblcLt15552H5DEAAAAAAAAAAICDZ1i1h0+0cJ9//rn+9Kc/he8vXLhQXq9XQ4cOldvtliSdddZZ+v3vfx/eZ8GCBRozZoxKS0s1fPhw9e7dW3PnzlV+fr4GDhyo2bNnKz09vUY7lmXp8ssv11tvvaW0tDSddtppKi0t1YwZM2Sapp5//nldf/31B/VYRo0aJUnKzs4+qHoAAAAAQOIcAwAAADgQR3N3IFo7d+7Uzz//XKd88eLF4dtHHXVUjW3Dhw/XokWL9Oijj+rbb7/VsmXL1LlzZ91///36/e9/r5SUlDr1GYahqVOn6qSTTtILL7ygL7/8Ui6XS2PHjtWDDz6oU089NeaPDQAAAAAAAAAANJ3DbqTIkYRvcQEAAACIJc4xAAAAgP074tcUAQAAAAAAAAAAkAhFAAAAAAAAAABAnCAUAQAAAAAAAAAAcYFQBAAAAAAAAAAAxAVCEQAAAAAAAAAAEBcIRQAAAAAAAAAAQFwgFAEAAAAAAAAAAHGBUAQAAAAAAAAAAMQFQhEAAAAAAAAAABAXCEUAAAAAAAAAAEBcIBQBAAAAAAAAAABxgVAEAAAAAAAAAADEBUIRAAAAAAAAAAAQFwhFAAAAAAAAAABAXCAUAQAAAAAAAAAAcYFQBAAAAAAAAAAAxAVCEQAAAAAAAAAAEBcIRQAAAAAAAAAAQFwgFAEAAAAAAAAAAHGBUAQAAAAAAAAAAMQFQhEAAAAAAAAAABAXCEUAAAAAAAAAAEBcIBQBAAAAAAAAAABxgVAEAAAAAAAAAADEBUIRAAAAAAAAAAAQFwhFAAAAAAAAAABAXCAUAQAAAAAAAAAAcYFQBAAAAAAAAAAAxAVCEQAAAAAAAAAAEBcIRQAAAAAAAAAAQFwgFAEAAAAAAAAAAHGBUAQAAAAAAAAAAMQFQhEAAAAAAAAAABAXCEUAAAAAAAAAAEBcIBQBAAAAAAAAAABxgVAEAAAAAAAAAADEBUIRAAAAAAAAAAAQFwhFAAAAAAAAAABAXCAUAQAAAAAAAAAAcYFQBAAAAAAAAAAAxAVCEQAAAAAAAAAAEBcIRQAAAAAAAAAAQFwgFAEAAAAAAAAAAHGBUAQAAAAAAAAAAMQFQhEAAAAAAAAAABAXHLGopLy8XLNmzdKsWbO0ePFi7dixQ3v37lVGRobatWunY445RmPGjNGYMWOUmJgYiyYBAAAAAAAAAACiclChyMqVKzVlyhRNnTpVJSUlsiyrxvaNGzdKkr799lv97W9/U2pqqq644grdcsstOvroow+maQAAAAAAAAAAgKg0KhTZunWrHn74Yb3xxhsyTVMdO3bUWWedpZEjR6p///5q3bq10tLSVFRUpD179mjlypWaO3euvv/+ez3zzDN69tlndeWVV2ry5Mnq2LFjrB8TAAAAAAAAAABAHYZVe3hHAyQnJ8vr9eqyyy7TVVddpTFjxsgwjAMeZ1mWZsyYoddff13vvPOO3G63SkpKGtXxI8GoUaMkSdnZ2c3cEwAAAABHAs4xAAAAgP1r1ELrF198sdasWaNXX31VY8eObVAgIkmGYei0007T66+/rjVr1mjSpEmNaR4AAAAAAAAAACBqjZo+6+WXXz7ohnv06BGTetB4npIiSZIrNb2ZewIAAAAAAAAAQNNr1EgRHP48JUWa/cjtmv3I7eFwBAAAAEcmT0kR//MBAAAAgJo4FKmsrNSKFSu0cuVKVVVVNWVTiEIoEClav0ZF69cQjAAAABzBSgsLNf2+m/XlPTcpe3GOvH6zubsEAAAAAM2mUdNnHYjX69XDDz+sp59+Wh6PR5KUkJCge++9V3/4wx9kszFA5VC7/vbfyef3yen36oSdy5TmLdWc9FRJ0gnr1+jNqy7QnLaD5LU75bA79OLTTzRzj+NT6HVqCF6n5sVrdXjgdQIQj0Lvfaak3f50Tdw5Rx29uyVJC/54l+5oe4bS7BXqqJ1y8d4HAAAAIM40SShy991367nnntNll12mESNGqLy8XG+//bYmT54sv9+vyZMnN0Wz2A+f3yfDU6lRu1cow1emRSlJ+qBta0lSit+voaWlOnbnSs1p1U8+V0Iz9zZ++fw++Xz+5u4GGoDX6vDA6wQgHvn8Pnl8fm0xOmjSzlnK8u7W7NCXYYp26cqdX+mldr9SiZGgfv5NzdxbAAAAADi0mmTIxquvvqrf//73eu2113Trrbfq3nvv1dy5c9W3b1+98sorTdEkDsDp9+rk3SuU6SuTKWl6ZkZ429eZGTIltfMWaeSeXNn93ubqJgAAAGKgUBmatHOWOnl3aUnwyzAftG2tpSlJ6uTdpet2fCqP5dJWtWnurgIAAADAIdWoUOSMM87Q+vXrI26rrKxUWVmZjjnmmBrlTqdT/fv31549exrTJA6Cp6RIJ+xcpkxfmSRpSUqStrtd4e3b3S4tSEmWJSnLW6jjd65SVfHe5uksajBlqNhIVqGRpmIjWcwADgAA9seyLBl+n87eNVedvLvq/TJMKBgp9iexxggAAACAuNKo6bOWLFmigQMH6oEHHtDvfvc7uVz7PmBPSEhQr1699Mwzz2jcuHFKTk6WJC1cuFAzZszQgAEDYtNzNEhoUfUMb6kk1TkxDnk7q63esdoo0TSVYJp66XfXq9DRSpbcsskRuBgOOQyHMtMy5Xa4leBMUKIzQYnuBFWVVWnv7iLZDbscNkN2w5LNkOyG6ty2G5LNkIYO6KtjhwyQ22mXy2GTy2GXy2HX62+9p4ry8vB+hnHgx3nf7TfIXS3okaTlK3P00RffRPV8HTP4aJ11+tg65S+9MU3btu+Mqq5brr1MrVtl1ijL37RFb7z7Ub3H+P2BaX5MSQW2dtphay2f4Qxvd1hetTN3K8vcIdNvavLfnomqT1dcfJ66delUo2z3nkI9+/JbUdXToX1bXXfFr+uUfz59phYtXRlVXedNHK+BR/etUVZV5dFfn34hqnpSk5N0xy3X1Cn/Yc4v+mHOvKjqOm30CRp53DF1yv/y1PPyeAMjqUKvVWMsXLJCX3wzK6pjRhw7VOPHnFinfMpLU7WnsCiquu646SqlpqbUKMtZt0HTPvwiqnoGHNVH5599ep3yN979UPmbtkZV17WXT1LHrHY1yrYW7NDLU9+Lqp5uXTrqiovPj+qYEL/fX+N36tfnT1TfXj1q7FNSUqqn/vNaVPW2ykzXrdddXqf8m1k/6ef5i6Oqa+L4MRo2pO7f0WjfC1xOp+6/48Y65XPnLdK338+Jqq5TTjhOp5xwfJ3yp559RSVl5VHVdaS8l0fSp2c3XXzB2XXK3/3gM+Wuz4+qLt7LG6Yh7+UN9ci9v61TdrDv5R6fXxUenyqqfHrxrQ+0e2+pfKbkswz5LMln1nMdvN2pUyd5/JYqPD6VB+spraiSKkp17c6c8Boikb4MszQlSUNLy9XJG5hKa/6yCRo1tG/EPgMAAADAkaZRoUhOTo4efvhhTZ48WW+++aaefvppTZgwIbz9z3/+s37961+rY8eO6t+/v8rKyrRq1SoZhqHHHnssZp3H/oUCkaL1a8JltU+Mq7MMQ+V2u8rt9mBJ4IPW2h/9lhTvp1HLUODHyiHDCl7LKYVvOyQrUDZj1h5p5vzA9mr7B/ZLDldpWKYMWbLJkiEzeG3JJlOGFSjf8PxMJbqdwWDFJrfTrt27dmv52rJ9+4bqsKrfr1m3PaW1+hfslcsZCGjcwev1eZuVv2lzNE9/xA9cysortDpn3X6PMyWts3dXkS2tzjaf4dRWe5bKjCT1NvMPWFek9iP1M9p6KiurIpYXbN8ZdV3FJ4+sU2ZaZtT1ZKTXfb4kaeeuPVHXNWxw5PA2Z+0GVVZFfuzR2FtUHHWfenTrHLF8/YZNKtgR3Ye8vgiBTmlpedR9ysyI/Jznb9oadV1VEZ7XqqqqqOs5GJalGu2Vltb9QN/n90fdp6x2bSOW79i5K+q6Tjh+WMTyaOtJcLsjlu8pLIq6rv59e0Usz12fr71F+/uDUZdp1f22enFJSdR96tA+8nO+IX+z8jYemvfy2hITIj/nm7cW8F5eTUt7LzclmbKpoLAsHDwEggyvflm+RbNz98g0bDJlk1+Ba1OGTMNe7b5NphHY/vGWtXroszxVVPnkM61arSXKkiXJUuC/L1OSX5ZMyah535Bfm3M3y2l5ZZdHDssrm7xKsqrUybNdP2VY8hpt5LEZWpOUWOdxfZ2ZocGl5bIpMGJkx5T/lefv/5ErNb2hTy8AAAAAHLYaFYqkpqbqX//6l6655hrdeuutmjhxoi644AI9+eST6tSpky644AItWLBAf/3rX7Vq1Sq53W5dfvnluu+++xgp0ozqGyUSU4YlySvJG8hHFDi1j4ol7QtWgoFJMEyRnNWClcB9WQ79uGZFYFuNYMWQ7O3qbSaS1csq9dqyjyJsaSXDkVktlKkezlgyrLrlT3y0RBmpydXCFZuKiopUYGsTIZzZd+xuIyMQiFiWLEMybRslSTaza+AxWZaKbGkqsNqqg7Uj2mcXTcSUoVIjSX7ZZZdfKVZZ0yzaBACoIRRcVL/4DZtW76iQfVFeeDRGKNTYYLaVx2bWDTMMm/yyB0IN2eSXEfxfxtToR96Q5JdlmAqFE5JfltMlGaH7nkCAUS3EsKqFGzJMWaZfMv0ybH4ZNp8MhfYzZRmmTFlSA0bISoH/tiojlG9JTDrgsdVHi0iSuW2DZj9yu06a/DTBCAAAAIAjXqNCkZBjjjlG2dnZeuGFF/TQQw+pf//++sMf/qA777xTQ4YM0dSpU2PVTzSCKzVdJ01+OjxaZH+jREL6lZUr2TRV7EjQ9qRW8sqU3/LJL59M+Q5Nxw1J8knyyTICp/sHF6zUDFP2hSzBsmqjV6rvHw5WQlUahizZI6/rEeEDjOlLt0Tum71jwx6DYci0bZPPEZjCxOFzyW52CM8ntsXWXrud7eSwKTg92b6pyerclmS3SW/O3aS2OcWBoMZhl9Nhk89TJWeHfvs/vtbtTu3bBOYsrzW3WVb7tjqqnm+N1yctNbVOmc2wRV1PanLkD4HatmkVdV2tMiN/INS3d4/wt8bX5K6TZTVsqrPaMtLTou5Tu7aRF8Lt2aOLMuoZsVEfR3g02D4pKUlR96ljVvuI5d26NPBnvBp3hJELbrc76j41pu0Qw5D69dnXXkpK3Z8ph90es5+ndm3bRF1Xfd+ij7Yel9MZsbxVZnrUdbVt0ypieZ+e3aKePstm1I0S01JTo+5TVj0jRXp066yEekZs1CfSc5WclBh1nzp3zKq3vKKeERv1SY7wzX+X0xl1n+obUdMU7+VmtWmf9ndd5nTpmS8Wqzw4IqO8yqsKj0+bC0q0PaWffJYpr2XKZ5ryWZb8ll9mMIiwgsFDaJTFip+zpZ9/qhtQ2P2So3pgERyBYdS8H6iraUT9f00TqD5aBAAAAADiiWFZVkzOy/bs2aP77rtPr776qgYOHKgpU6boxBPrzn+PfUaNGiVJys7ObtJ2PCVFeuOqC/RCVtoBQ5H2VR79dvNunffKB0ptXfNDWMuyVFZVqZLKcpVUlAeuq90urihXSWXFvrKKchVVlNXYt6yqQmZsfuQODcuu6iNWIo1eCYUp1Uev1Byx0riPGyxZ8jpmy7IF1oMxzBQ5fSfVCGqak2FIbse+acacwWnLQoGLy2EPTkNWrTx4HZrmrPoomkAd9lp1RCgP7R+sx+mw1QlnDoWrb71PHp+/3qnOQtLNYvUzNun1KX89hL1DyNW33iefr2Hrvzgcdr3K64Q45jfNGiMqQmtVlFd5992v8qm0skollZUqqahQaWWVyqoqVVZZpQqPR+WeKlV4qlTp9arK61GV3yuP1yOf6QuGDNWmhKoWYoTvB0dZ1Lgvf3AkKg7EYVpyWqacViDyKXPs//tPVxbs0NDScqX37McokSPIoTrHAAAAAA5XBzVSpLpWrVrppZde0vXXX69bbrlFp5xyiq688kr99a9/VZs2kb/hjEPDlZquj7J6a7t77wH33e52aW2yXSue/bOOu3+y7K5936g1DEMpCYlKSUhUh4zWjeqLZVkq91TtJ1QprxW6VOy7XW2732y6b2/WYASnxzD2fYs3+o9l7PtCEqvm+il111vZN3rFNPaEAxFJsmylMm0FgdEiLYBlSZVevyq9fqnu9PaHlKtWION01gxVXLVDGkfNgKX2GjKhY+uUO/eFOVVyaputbYOmOttq8R4IIDb8pqkKj18VVV6VV3lVXFGpwrIKFVdUqKSiQiUVlSqpDIQUpVVVKq+sVJknEFhUevddPF6vPH6vPD6vvH6ffKZXpuWvOd1TxIDCjD6gsAUvccSwLDktS47gtdOMdD8QXkTa7pKU4HTJ7XIrwZWgxIQEJboTlZiYpKTEZCUlJyspKVVJKalKSUlTcmqGklPT5EpJlWVaev+u6/V8VrrKDvCf/teZGepWZdNZBCIAAAAA4shBhSKLFi3S4sWLtWfPHnXr1k1jxozRqFGjtHDhQv3rX//So48+qk8++USPP/64brrpplj1GVHym6bWO8savP/XmRka/PMP8v7fXRr58F/lTEo+8EENZBiGkt0JSnYnKEuRp1w5EMuyVOGpUklluUorK1QcceRKRcTgpfptb4TFpptGaEqOqvA0W439vqvPvkQ++2oZskuWTamJCXI5nMHRKPsulmVIlk2maZNpGTJNQ/7QxS9ZVnDf4HXgePu+OixbnToD+7WMUSq1eXymPD5TgRnWDxGjb+Apkw441dlWtdUzXyxWq5QEpSe7lZHsVnqSW+nJLmUkJyjZ7WiW0S4AYsc0TVV6ParyeVVeVaWi8goVlQfCisAlGFRUVgbCiqqq8KiKUGDh8XlV5Qtce3xe+fw++czAxbT88pu+CKFFjB6AodjV1cLYagUP+8IHU06zWlBRa5/Q/fC+kbZbVnB0xr6LyzCUkJQiV3KKHEkpcqYky5mUIkdyipxJKXIm174fvARvO5JSZHe7D+rvwodZfRr8ZZiPsnrrWgIRAAAAAHGkUaFIUVGRLr/8cn3xxReqPvtWSkqKJk+erP/5n//RnXfeqUsuuUR33XWXbrnlFr3yyit69tlndcwxx8Ss82iY6cvmqcLwhr/YaavnI3lThmyytMvl1LLkJA1ZukA/PXKbRj36pNxpLedk2TAMJbkTlOROUPtGdsuyLFV5vYERKtVDk/0GKXWDFo//EK2zEmJYkioDr6AhFVeVSNFNRb/vg/xGcNjsstscstvsshsO2Wx22Qy7bLLJMOw1wxXLJsmQVSucCQQ0kt9vyLSMwP41wpnQxX7YhDOWLPlta8P3/ba1splZ4X6ahk1Pf7643uMdNqNWWLLvdkayK2J5erKbMKUBHPaG/5mLZl+0bF6/T1Veb3hURJUvMJVT+L7XGwzXAyMrSoNhRVlVpcqqAiFF4BKcBsoXOMbj98rnD46s8HvlM/0yLV9whMUhGMF4BPy62xswSqLO/WB4sb8govboC0e1eqL5s2fY7TWCinBwEQwzHEm1A4zkmAcaB8tvmtrgaviXYda7ymSapmy2OBvOAwAAACBuNWpNkWuvvVavvvqqrr76at1www1q1aqVVq1apT/84Q9avny5vvvuO40ePTq8/4wZM/Tb3/5W69at080336ynn346pg/icNVc8/16Sor04yO3qXh9jiTJ2amnxv6/KUrOyNTOpfM1d/J98lXsWxw3tUsPnfinfymxdbtD2s/DQZXXEyFUqagZqkSaKqxawOLxHcLRDUcIu80uRzCcsRn28LVN9mrhjC0Y8wUCGssKjKAJBDOBkMYXHDljmoZqjpypNmqmxsiZmuWh0MNv2yafY3GNPjp8Q5t8qjOn3ab0JFetECUQmIRuB8pdNcqT4jhM8Zum3pzzjSTp8hPG8yFgE7MsSx6/LxBIeALhRGWNcMKjymB4UeX1qMpXM7io9HpUWlWlsspKlXuqVF4VCio81cIKj7zBoCIUVlgtYhnrli8wcWNg0kaXpWrhgimXacrp98nh99c/SqLWKAtH8Ni6QcW+7U35G2c4HPsCjaTkwzLQiIUvl/ysO9/4l6TAyFTL56vzO6rOl7UAAQAASURBVGHIkOFwhHO2J6/4H505ZMSh7SiaDGuKAAAAAPvXqFCkVatWOvroozV79uwa5WvXrlXfvn1155136h//+EeNbV6vV3/5y1/0xBNPqLS0VGjeExZPSZFmP3K7JNVZWLMwZ6XmPHqnPCVF4bKkdh104p+eVkrHLoe8r0e6Kp+3RoDy1dJ5enbGh/s9Jt3dTsf27ipvcD74Km/w2ueVx+cLTr8S+CDS4/PJZx6qqcLijBWabybCFDaWTbLSZWhfSBMKXeqMiok0Uibi6JlgmVW3LJoRNE6HTRlJ1UMU1/4DlZQEpSe5lOg6/MOUj+bP1l1vBoL5O8Zfpd9OOENOe3wEI6ZphkOJ6qMoKquFEVXemiMqQqMsqocYdY4Njroo93hU6a0K1OELHOeN8GEsIrAkBYNdm7FvNJ7TZpPLsMlls8ltGHJLchsKXFuW3DLlNv1y+f1y+b1y+bxyej2yV1WGL/WFEw7LalEDT/4/e/cdHlWVuHH8vVPSQ0hoCb33Jh1RBFRQXMsq/tRd++rae0ddK/Zd117XteJaWMsiKoo0qVKlIyUIhEBCep9yf38kmWRIIQk3mSTz/TxPnpk5t5yTDExy7nvPObUNNPxGcZQcYwtp+oFGfSj9my9z1zZJYlH1IEAoAgAAAFSvTqFIdHS0xo0bp++++86vPDk5We3bt9cNN9xQ5WiQPXv2qEuXLnVrbTMT6A5LaehRWac46/ddWvK3m1VwOMVXFtoyTuMefVEx3Xo1WBuDjcfr1enP3KXfDu6rdr+e7Tro+7ufq/Fd7h6vt5LwpPixsJLyyoIV3zGVnKPq8vLHFzXgOi5ByiwZ8aIjwhfzyBEudQhpZJPDZldUWJiiwkIVHR6mmPAwtYgIV2xkuFpGFj+2io5QXHSk2kRHKrZkDZUwpz3gFypdHq/e+H69XvjxFbnNbEmS4Y1SQvip+vNJ/XT15EENGo64PZ4qAociv1EQvvChqPIpoI4MLwqr2caotBoqndavZHRY6f8hw7DLbhRPIeiwO+S0O+W0OxTicCrUEVK8KLbDqfCQUIU7HYq0GYqwGYo0TIUbpiJMr8JNt8K9HoV5XAp1FynEVSBnUYFs+fly5eXIlZvje/QU5Af4B1FzRwYazsjoksCibCRGtetpEGjUu+puhkHzE+g+BgAAANDY1WkC9VNOOUVff/21nnrqKV155ZVq2bKltm7dqjvvvFOGYejkk0+u8lgCkcajug5xi87dNf7pN7XkwZuUe6D4An1hRpoW33edxj70D7XqN7ihmhlUvl2//KiBiCTtOLhfc9av0B+OG1uj89pttuILdSGhx9rEY+L1elXkdqvI4yq5m7x8EHP0ES9lr2sezBS63SpyFRU/uksfm+nFYcNU8eLLZeFT6foz5dX1vn23pIIiKbVIUlYNDigJaQyVTG9ms8ths8tpd8jpcCjU4VSoM0RhTqfCQ0IUERKiiNBQRYUVf0WEhCrE4VCo06kQR+mXQyH2suehpWUOZ8nzcvuVPLcZNt3+zmLN37JKbkd2WfNsOTqYu0svzM7Tqp379fifx8jjLf53VlBUi5ESftuPCC6qOJbRWzVk+k9bZ5hlYYVvqrty+9gMu5w2p5yOsrCi+N9YiMKdxf/GwkNDFREapijfv7WwkpAvQi3CwxQZYleoWaQQd6GcnkLZXQVyFOXLW5gnd16u3Lk5KsrNljsvR67cXLny0uTKyCn3+uiBhkdSXslXoNkczioCDAKN5iQkOkYnPP6S7zkAAAAABLM6jRTZt2+fTj/9dG3atKlCJ/i6667Tyy+/bFkDm7OmcBdXQfphLfnbzcpKLFtE2h4aptHTn1a7YWMC2LLmp6ajREr1atdR3939LGsi1EHpOgeVBik1DmaKtD7xN83ftr7auiLsUerSrpVcHndxIHREkMPomQZkqpJpziopQ0WmVNnoieoCCt/aPGbp69Jj7Qp1OhVWMroirCQMiwgJVWRoyVfJiKSosOJp28JDHAoPdSii9DHUqfAQhyJCHb7tEaEOOQxT7ry8koCibNSFOy/3iNdlAcaRr5vSCI1aBxqR0RX2sQc4LAdgvabQxwAAAAACqU4jRTp27Khff/1Vs2bN0rp165Senq7OnTtr6tSpGjRokNVtPGYLFizQxIkTj7rfxIkT9dNPP/leH+2ux2XLlmnMmOYdDITFttKJT76uZY/errQtv0qSPIUFWvbYHRp5x6PqcELVo4JQO3M3/KI9qckKsdfsv+We1GR9v+EXFkatA8MwikcpOJyKDqvbOfIz03XKd7OOekG9pTtX//3rcwqLia10u9frlcvj9o1gKZ2+rOiIUTRlU5iV7OcpC3DKwpaybaX7++3jqXz/8vU065CmsveqKQYivmnSahpGlAsvzHKhRmWvfVNF2UpGVoQWBxahIYoIdZYLIZwl4URZUBEe4iwXWpQGFU6/0CIi1KEQR+XTqXk9brlyc48INLKKA43sHLkO+gcYWXk5OtycAo3IaP+1NHyhRuQRrwk0AAAAAAA4FnUKRaTii4rTpk3TtGnTrGxPvYiPj9dll11W5fYvvvhCWVlZOumkkypsi4yMrPJ7bNOmjWVtbMxCoqI17tEXtfLJ+3RwTfEdZ6bbrZXPPqDjcnPUdcrZAW5h83D6kNE6fchHgW4GaqAoO1Mv/O1GJdmOPtAuyTD1wt9u1G2Pv1rplCU2m02htuIpfhqDqkKa0hEylYc0RcotLFJGbr6y8guUnV+g7IIC5RYUL76dV1io/KKyBb6Lgxu3XB63TNOj4oXqvTLlLX5e/sto5It0+9ZlKQkgajx6omzURNlaLnaVBRRloYVfqOEbcSEZhhTu9B854RtNUW5ERVloUT6cqHz/yJLy0FquA1Mx0EiTK61shEZ2Xo7ScktCjdLAoyTMKD3GU1hgzXvSAGzOkLLFvytbHJxAAwAAAACARqtO02c1J3v37lXXrl1lmqZ27typbt26+bYZhqEuXbooMTGxXupuakPbvS6XVv39Ie1fMs+vfMAVN6r3uZcEqFVAwyrKztSiB27UA64cHQytWZDRrrBID+QUqscZ0xQWEytHWLjsoeFyhIfLHhYuR7kve1i47KFhQTE/v2mayi9yKzOvSBk5BcWPuYXKyC1UZl6hMnMLlZ5ToPTcPGXkFigjN1+Z+fnKyiuQy+vWkQGKWRKuHFnmNTJl2g9U2xbD0152s6X8RkuUHz0huwzT/3VxQHH098lvmidfCFH5SIvyIyqK93ceEXI4FFFyrFUL2HvdbrnycuXKzS4LLEpGXrjLTTnlKllPI+gCjajoCutp2BtJiAkAlWlqfQwAAACgodVppMiPP/6oU0455Zgrt+o8x+KDDz6Q1+vV+PHj/QIRVGRzOjXyrsfkjIpS4vdf+co3/ftlubKz1P/S64PiQi6CV1F2pn5+4Cb9fHCfUuPbyF7DTDk1xKlVRRlyffhGjesqDkvC5AiL8AUljvAIOUoe7eUfw0oCllD/R0doSehSWhYWLpujzgMELWcYRsmUTE4lxEbW+DjTNJVX6PYFJ8UhSsVAJSO3UHtTM7Up5ZsaNCZLNs9gtWkRrtjIML+RFpWtXVHTkRlhTodstvr7XPS63b6ppoIy0IiK9o28cJQLOPxfE2gAAAAAAIAydbo6NnnyZJ100kl64IEHdPLJtV9TYu7cuZoxY4Z+/vlneQI8f/37778vSdVOr4Uyht2uoTfcJ2dUC/026wNf+fbP31dRTraGXnuXDLs9gC0E6t+Q3DwN2bmnXuvwFOTLU5CvQqVbel6bwyl7+cCk3AiV4sfiIKYskCn/GF5JWfG+tpDQBgtFDcNQZJhTkWFOtY+LqnbfL1f9rNs+yjnqOU1bjiIj0zT/8dvktNusamq1ah1o5GaXjOgoO8ZTVNggbbWCLSS0ygXBCTQAAAAAAEBDqVMo8uKLL+qRRx7R5MmT1aFDB/35z3/WhAkTNGrUKMXGVlxMOC0tTStXrtRPP/2kjz/+WElJSYqLi9OLL754zN/AsVixYoW2bdumiIgInX/++ZXuk5ubqyeeeEJ79uxRWFiYBgwYoLPPPlvt2rVr4NY2HoZhaODlNyokqoU2vfeKrzzxuy/kys3WiNsels3pDGALgfoREh2jEx5/ST8/cJMyd20LdHPqxOt2yZvtkis7y9oT22xyhIZVEbRULKt+e7nQJTS8zkGrx+vVK3M/r/m34Nwhew1zncoDjZwjynKrfd2cA42QyOgK62nwewEAAAAAADQGdV5TJDMzU08//bTefvttpaam+u4Qbt26teLi4hQdHa2srCylpaXp8OHDkoqnPGndurWuvvpq3X333YqJqbjocEO64YYb9Oqrr+riiy/WBx98UGF7VXc9h4WF6YknntBtt91Wo3pK5/U90saNG9WvXz/99NNPNW90I7Nv3mxt+dc/pXL/jFoNHaUhtz4ke2hY4BoG1BPTNLX7y4+089N/1/iY6K49ddw9T8rmDJGnMF+egoLKHwsL5C4ofvT4Hivbt9z2JjT1UV3ZnCG+KcTsJcFL+UeH73WY7KHhvsfFh/bp6Q3LVNNfcoakO3sM0pgWreTOy5U7P7f4sfSr3GtvUwo0nCFyRESW+4oqex5e/Ogsvy088oj9I2VzEGgAOHZRUdWP7kPtNOc+BgAAAFATde1jHPNC60VFRfr888/1v//9T4sWLdKBAxUXs23fvr3Gjx+vM888U+edd55CQgI//UVRUZESEhKUlpamH374odK1TS699FJddNFFGjx4sGJjY7Vz5069+eabeu211+TxePT666/rmmuuOWpdzb3Dkrxsvja+8pRMj9tX1rLPQA29a4ackXR+0Xx4CvK16c2/6+Cy+TU+JrprTw2//1k5o1rUS5tMr1feosKKYYpfsFKxrCb7mwGe3hDFIzSODDAINAA0VYQi1mrufQwAAADgaAIWihwpNTVVhw4dUkZGhlq2bKm2bduqdevWVlZhiVmzZmnatGnq2LGj9uzZI5ut5nPIv/7667ruuusUFxenpKQkhYaG1qkNpR2ZZcuW1en4xuTg6mVa8cQ9ftPBxHTvreMf/qfCYlsFsGWANXKSfteKJ+5V1p6dfuXOqGi5crIrPSamex+d8PhLCokO7Ki4uvK6XHIX5MldUCB3QZ48vsf8SsqKgxZ3yVoo5R+PLGtK00YdC3tIaMk0U2XTS1X72m9qquLtTDkFAKit5tTHAAAAAOqD5aFIU3H22Wfr66+/1n333acnnniiVsd6vV7Fx8crJSVFCxYs0EknnVSnNjS3Dsvhzeu17NHb5cotW9Q4sn0nnfDYS4pomxDAlgHH5sDKxVr9j4f9/m0bNrsGXnGTOk06XUsevLnCGiNNPRCpT6bHI3fJ6BR3fp48hfly5/sHKKXb3KX7FBTIXZgvT/my8o8lx8rrtaSNNmeInFHRVQQWR66t4b8YOIEGACCQmlsfAwAAALBanRZab+pSUlL07bffSpIuu+yyWh9vs9nUs2dPpaSkKCkpyermNVmt+g/RiU++piV/u1mFGemSpNykvVp0z1817tGXFN2pa2AbCNSS6fFoy8dva9sn7/iVh7aM1ai7n1DrQcMkqcLi6wQi1TPs9uJgISJSsnAkmWma8rqK5M7PV/7hQ1r13IPK3ptYq3O06NJT4x57SWGxcZa1CwAAAAAAAI1HzeeMakY+/vhjuVwujRkzRn369KnTOdLTiy/6R0ZGWtm0Ji+mW2+Nf/pNhbeJ95Xlpx7SonuvUfqOLQFsGVA7RdmZWvbYHRUCkdg+AzXx+fd8gYgkhUTH6ITHX1JM9z4EIgFkGIbsIaEKjWmplt2LP4tiutf8Mz6mex+d+OSrBCIAAAAAAADNWFCGIu+9956kuo0SkYoXL9y6daskafjw4Za1q7mIat9ZJz3zpqI6dvWVFWVl6Ofp1ytlw+rANQyooYxd2zX/tst1cLX/tBPdTj9PJz75msJbt6twTGkwQiDSeJQPq46GMAsAAAAAACA4BF0osmnTJq1Zs0ahoaG64IILqtzvvffe05o1ayqUr1mzRtOmTZMknXfeeerQoUO9tbUpC2/dTuOfekMte/b1lbnz87T0oVt1YOXiALYMqN7v87/VwruuUt7BsqnxbM4QDbvlQQ29/m7ZnSFVHhsSHcNF9UamJsEIgQgAAAAAAEDwCLpQpHSUyFlnnaXY2Ngq9/viiy80fPhw9e3bV+eee64uvPBCjRgxQiNHjtS2bds0fPhwvfnmmw3V7CYpNKalTpjxiloPLJtmyOsq0ooZ9+j3+d8GsGVARV6XS+vfeE6r//GwvEWFvvKItgka/8yb6nLKHwLYOhyL6oIRAhEAAAAAAIDgElShiMfj0UcffSTp6FNnXXrppbroootkGIYWLFigWbNmaffu3Ro/frxeffVVLV26VHFxzDt/NM6IKB3/8POKH3Wir8z0erT6Hw9r1zefBbBlQJn8wylafP/12jXb/99k26GjNOH59xTbs1+AWgarVBaMEIgAAAAAAAAEH8M0TTPQjQhWY8eOlSQtW7bsKHs2fV63W2teeEx7F3znV97v4mvU5/+ukGEYAWoZgl3qpnVa+fR0FaYf9ivvff7l6v/nv8qw2wPUMtSHouxM/fzATZJEIAIAaJaCqY8BAAAA1IXDipO8//77uuCCCxQaGmrF6dAM2RwODb/tITmjWmjX7E995Vs+fENF2VkadOXNMmxBNXAJAWaapnbN/kwb/vVPmR6Pr9wRHqHhtz2s9mNPCmDrUF9KR4yUPgcAAAAAAEBwseQq9OWXX66OHTvqjjvu0Pbt2604JZohw2bT4L/err4XXeVXvvOrj7Xmxcfl9bgD1DIEG3dBgVb/42H9+ubf/QKR6E7dNOEf7xKINHMh0TEEIgAAAAAAAEHKklDkvPPOU1ZWlp5//nn169dPJ598sj777DO53Vzkhj/DMNTvT1dr0NW3+ZX/Pu8brXz6fnnKLXAN1IecA/u08K6/VJjKrf24STrpuX8pumOXALUMAAAAAAAAQH2zJBT57LPPtHfvXs2YMUNdunTR/PnzdeGFF6pjx46aPn26EhMTragGzUjPsy7UsFselMpNmXVg2QIte/QOufJyA9gyNGfJq5ZowW2XKytxR1mhzaaBV9ykUfc8IWdEZOAaBwAAAAAAAKDe1ctC699//73eeOMNzZ49W263WzabTZMnT9a1116rP/zhD7KxdoQkFkGUpKRlC/XLM/fL63b5ymJ7D9DYh55XaAumt4E1TK9XWz95R1s/flsq95EX0qKlRt09Q22GjAhg6wAAAKxDHwNoPjxerz5a+oMk6eLjT+VaCgAAFqmX36hTpkzRf//7X/3+++967LHH1KlTJ33//ff64x//qC5duujRRx/VwYMH66NqNDHtx56ksQ8/L3tYuK8sffsmLb7vWuUfTglgy9BcFOVkadljd2rrzLf8ApHY3gM08Z/vE4gAAAAAaJS+Xb9cD816Rw/Nekdz1q8IdHMAAGg26vU2g/j4eI0bN04jR46UaZoyTVP79+/Xww8/rG7duum+++6Tp9wixwhObYeM1AkzXpEzuoWvLPv3XVp0z1+Vc2BfAFuGpi5z929acPsVOrhqiV951ynn6MSnXldEm3YBahkAAAAAVM3j9erF72f5Xr/4/efyer0BbBEAAM1HvYQi6enpfouuf/755+rZs6eee+45rVu3Tvfee6/Cw8P1zDPPaPr06fXRBDQxcb0HaPyTbygsro2vLO9gkhbd81dlll//AaihvQu+18I7/6LccsGazRmi4266X8fdeJ/szpAAtg4AAAAAqvbt+uX67WBZX+a3g/sYLQIAgEUsXVNkyZIlev311zVr1iwVFhbKZrPpjDPO0PXXX6/Jkyf77Xvo0CENHz5cXq9X+/fvt6oJTQrz/VaUm5ykJX+7ye9CtjOqhY5/6B+K6zsogC1DU+F1u7Xx3y9q59ef+JWHt26n0dOfVmyvfgFqGQAAQP2jjwE0bUVut7Ym7dHV/3pWh7LS/bbFRkbrlinT1CG2jRJaximhZSvFRkbLMIwAtRYAgKbJklDkpZde0ptvvqnNmzfLNE21a9dOf/nLX3TNNdeoU6dOVR532WWX6cMPPwzaKbTosFSuIC1VSx66RVnlRojYQ8M05v5n1Pa40QFsGRq7gvTDWvn0dB3etM6vvM2QkRp512MKjYkNTMMAAAAaCH0MoOnIKyzQ1gO/a+O+3dpU8rU9ea9ctbhGEup0KiGmlRJatlJ8yzgltGztC0xKv1pGRBGcAABQjsOKk9xyyy2SpBNOOEHXX3+9zjvvPDmdzqMeN2jQII0fP96KJqAZCYtrrROffE3LHrldaVs3SJI8hQVa+ujtGnnnY+owblKAW4jG6PCWX7XyqekqSEvxK+913qXqf8k1stkt+bgDAAAAgFrLzMvRpv2JvvBj0/5E7TqUJO8x3qda6HIpMTVZianJVe4T5gxRfMtWat+yleJj4pQQ28ovSGnfsrViIiIJTgAAQcOSkSLXX3+9rr/+eg0cONCKNgUN7uKqnrsgXyueuEeH1pabN9Vm03E33qeup54VuIahUTFNU7vnzNKvbz8v0+32lTvCIzTslgcJ0QAAQFChjwEE3qHMdG3an6iN+3Zp8/5Ebdy3W/uOuHmrsQkPCS0OTMqNMEkoGXlSHJy0UotwghMAQLH8Ipeenf2F8os8Ouu48RrRM0FOe70sX14vLF1TBLVDh+XoPK4irfr7Q0pa8pNf+cArb1avP/45QK1CY+EpLNDaV5/W3p/m+JVHdeyq0dOfUotO3QLUMgAAgMCgjwE0HNM0tS8tpXj6q/2lU2AlKiU7o9bniggJVb/2XbTjUJIy83Kq3bdjXBvddcafdCgrTQfSD+tA5mEdyDis5Iw0HcpKP+bRJ9W1Mb5lKyWUjjZp2VrxMXFqH9tK8TGt1D62laLDIghOAKAZc3m8emvuBr09/0elu1ZKkhzuoWoX1V1/Gt9XV08e1CTCEUvmkyksLNTBgwcVGxur6OjoSvfJzs5Wenq64uPjFRISYkW1CAJ2Z4hG3fW41kY8pT0/fO0r3/jOiyrKzlL/S67lD64glZucpBVP3qPMXdv9ytuPnaBhtz4oZ0RUgFoGAAAAoLnxeL3adShJm/bv1sZ9u7V5X6I27U9UVn5urc/VMiJKAzp204AOXTWwYzf179hNXVvH69v1y3XT+y8c9fh9aSmyGYaumvCHCttcHrdSsjJ0IOPwEV9pxcFJ5mEdyspQXe6PzSsq1K5DSdp1KKnKfSJCQkum5SqZrqtk1En7cmueRIeF048HgCbI5fHqxjd/0oKNe+VybJZKsg+PbYdSsuL14uy1Wp+Yopf/OqnRByOWhCLPP/+87r//fs2fP7/KNULWrl2riRMn6plnntEdd9xhRbUIEobdruNumi5nVLR2fPGRr3z7Z+/KlZOtIdfeKcPWuP+jwVoH1yzXL88+KFdOVlmhzaYBl1ynXuddwh/YAAAAAOqs0O3Sbwf2auP+4pEfm/bt1tYDvyu/qLDW54qPiVP/kvBjQMlX+5atKvRZPF6vXvx+Vo3P++L3n2vqkNGyHdEXdtodah/bWu1jW1d5rMvj1qGsDB1ITy0OTDKLR5wkZx5WUsmIk5TsugcnOw8laWc1wUlkaFhxcFIyuuTIdU4SSkacNBVFbo/W7jqkzLwixUSEaFiPdo3+YiAA1MVbczdo4cZ98tqSZdrKRjWathx5bcmyexO0cOM+vT13g647fUgAW3p0lkyfdfzxxys5OVm7du2qdr9u3bqpY8eOWrx48bFW2SwwtL12TNPU9s/f0+b3X/Mr7zh+sobf9pBsDhbSbu5Mr1fbPntXWz56Uyr30RUSHaORdz+utkNHBbB1AAAAgUcfA6id3MICbdmfWLYI+v5E/Za8Vy6Pp9bn6tK6nQZ06KYBHbuWjATpptbRMTU69tv1K3TrBy/KlOT2emV6q75UY9gMOWw2vXDJzTp9yOhat7MmitxuHcpKL5mWqzgsKX1eOvKkLtOE1VRUaLhvIfjy65yULgwf3zIu4MFJ6RQyMxdtVWpWvq+8dYvwJjWFDADreL2mvGbJl++55PV65TWLr216TVMerynTNOX2elXkdsvt8cjl9qjI45HHW/zc7fHI5XHL5fHK7S1+Xfrl8njk9nrk8T33yuP1+PYrfu6Vp+RYj7f00SOPxyuP1yuP6ZHb45W35NFjeuT1Fm/zmmX7lJZ5vB7tPpQpt9cj0zgsGf6/Jw1vpJzuE2XIUOsW4Zr/+PmN+jPQkqvIO3bs0MiRI4+6X//+/bVmzRorqkQQMgxDfc6/XCFRLbTutWd8F8X3LZorV16uRt3zhBxhYQFuJeqLKzdHq55/WMkr/EPVlj37avR9TymibUKAWgYAAACgKcjIzSmb/qpkAfTdKQdqPSLCZhjqFd/RNwKkf4du6t+hq1qE1/0i/elDRuv0IR/p1W/X68XZa4+6/y1/OE6nD6m/u3BDHA51jGujjnFtqtynODhJU1L6YSVnpimpZORJcmbZWiep2Zl1qj+nMF+/Hdyn3w7uq3Kf6LBwxce0KlnfpHStE/91TqLCwutU/9GUTiGzcOM+mTLltf0uSbJ5Oys1K79JTSGDxsU0TZmm5DFNmV7/C+hlz1XJhffS5/IrN0uee0rO63teem7f84rHeXzPVXKhvOJF/fIX/I9sS/H34ZXXK7/zeb1e30V438V634X94ovvxRf1Td9zr+ktPq7kucfjLT63WXaMaZZd0C9uk1l8Ub9kP7Ncuamy/UyZZdtLthV/b16Z8kq+52bxc3klmTJLy0qeq+S5DNNXXvblPeJ16b4B+Wd2bKr4SDNtub7RIqlZ+Vqz86BG92681+osCUWysrIUE3P0ux9iYmKUnp5uRZUIYt1OP1fOyCit+sfDMkvu3jm4aomWPnyLxj74dzkjWUuiucnas1PLn7hHuUl7/cq7nHqmhlx7l+whoQFqGQAAAIDGxjRNHcpKL1kAPdG3APr+9JRanyvE4VTfhE6+kR8DOnZT34TOCrNwrVSv11RmXqEOpOfqvZ821eiYN77/VQfScuVw2GQYkt1myDAM2QyjkueqUG6zlTwaks1mK340irfbS7YVn7d4m2EzZC/Zbit9bis+3m5EKy68hVpHdNPQTqXnLd7P43EpIy9Lh3MzlJaTqdScdB3OyVBqdrpSstOVmp2u9LzsOv3csgvylV1wtOAkQgkla5kUjzaJq7DmSWRo7W+uLJ1CRpK8tmS5HZslSQ53iOze4ouAgZxCpvTC9VEvWJe70F3xwnvx8eXLK979XrG89AJ7VRf1/S7cV3oRXkdckD/iuMruxK/iOP/vw1tFmFDuZ1QhTCi+qG96dUSAcPR6y8KE8j/X0ovynpLzeX0X7L3e0gv3ZRfWiy+kS5VdUPe76O7bt5p9Si/MG9KRF+kr7ld63hrsc8S+1e9Xcr6mGARUxqjieRDz2HbI5o2XIUOZeUWBbk61LAlF4uPjtXHjxqPut3HjRrVuXfW8lkBNdRw/WY6ISK148j55S+Z1PbxpnRZPv17jHvmnQlvGBbiFsMq+xT9ozQuPy1NY4CuzOZwafM0d6jrlHNYPAQAAAIKYaZr6/fDB4rU/StYA2bhvtw7n1H6EQmRomPp36KoBHbqVLIDeVT3bdZDTXvtLJ6ZpKiu/SKlZ+b6vw9kFSsnK1+Hyz7OLX7urmS6rMgUujz5dur3W7Wocwku+2kuSQuSRVCjTyJdpFMhUgUyjQPJ7XreLa9kFecpOztP25KqDE5vhlNOIkNMeIactQqH2CIU4ohRqj1CoI1Jhjkg57SG+oEeS1u46JEklo0TK3gevbbvvgqAkvfrtei3dVry+ylHv4q/swnu5C/fVX3j3v+B/7BPlW+foF8klqeSOe6P6fSucq8L+3pI6vUfZr/TifVV371cWLFS9r3lE/ZXua1TxphglXwwoQiCZpf8QK/sq/sdpyCbTdEu2qtfXKr+2SEyEdTcP1AdLQpETTzxRM2fO1Jdffqlzzjmn0n2+/vprbdy4URdeeKEVVQKKHzFO4x59QcsevUPuvFxJUuaubVp07zUa9+hLimgbH+AW4lh4PW5tevcV7fhypl95WKs2Gn3fU4rrMzBALQMAAAAQCG6PR7sOJflNf7V5f6KyC/Jqfa64yGj1Lwk/StcA6dKqXYWFy8szTVM5BS5fsJGala/U7IKSwCPfL/BIzc6Xy+2t8lw1Fe4pvjks3958p4o2ZJcUIcOMKL5+XAlTHqkkIKk8OMmXDFed6veaLhWamSr0VhOkmU4ZZpgMhUlmWPFzW5jsZqa8trJ/f15bnhzaK7c6SZKKPC6t/C1JVV1QN/0umFdx0b22owKMcvtUGQbUdVRAyb/pYBwVgGaoZFScbDIMQ4Zhk82wyZBR/GjYSka+2XzbbIatZHRd6Zchm80u+xHb7Lbi53bDXu55cbm9ZH+7vXib77lRvM1ht8thK9vXYS959G0r2W4v99pul7P0td0mZ+m+drucDocctuLtztLtdkfxa4fDV5etZPRfhVGDJaMDC1wuDbnvernNqkMRqXi0SNvIbhrWo10DvY91Y8lC6+vWrdPIkSMVEhKihx9+WJdffrnatCmedzI1NVXvvvuuHn74YRUWFmrZsmUaMWLEMTe8OWARRGtk7NymJQ/doqLMdF9ZeOt2GvfYS4ru2CWALUNdFaQf1i/PPKDUjf5rELUeNEyj7p7BSCAAAIAq0MdAc1HoKtK2A3vLLYC+W1uS9qjQVfsL3wktW5VMf9XV95jQspUMw5BpmsotcCk1u0CHs/J9ozf8Ag9fAJKvIguCjpoK9xToL4f+J0n6V9szKwQjLcKdstttVU5nZJZbRyAYVB2c5Jd7XrfgpHYNMSVmNEAjZj/i4r3dVnLhvuQCvt3mX24vvQh/xGuH3V62zW4ruVhfdjG+9LndZiu+GG+zy+EofrTbive3lTzajzhfWbl/KFC+bb5goNz5KgsS/I+xy263+X4GqLnZa5fqpvdfqNG+Zw0+Ry9ccVE9t+jYWBKKSNLrr7+uG2+80bdAWekaI5mZxUm7YRh68cUXdf3111tRXbNAh8U62ft/15IHblR+6kFfWUiLlhr3yAtq2bNvAFuG2krbukErnrpPBYf95/vt+cc/a8Bl18tWh6HrAAAAwYI+BpqinIJ8bUnaUzLyo3gh9B3J++X2emp9rq5tEjSgQ1f1ju+sDi0TFBfVVi6Xo9rAo8BV+3rqym4zFBcdptbR4WrdIlytosPUpkW4WrUIV8vIUD35+Upl5hX5ApEOrlRJ0n5na79gpHWLcM1//PwaLeJdm4WbK679UPWaCzVdX8JTOv3T0daXqPK8la/3cOS6DRXLy31vJe1weVzKLchWTlG2cguzlVeUozxXjvKKcpTvylG+O0cuT/V3QaNpshmG/0V4e7mL7n4X/I+42H9kMGC3+118L74YX3aHv9/F92rPV9qOsvKy9vmfr7ryGp+vpN2lowCA2vB4vTr9mbuqXcOpvJ7tOuj7u59r1MGTZVcXr732Wg0ePFhPPPGEFixYoIyMDElSRESEJk6cqHvvvVfjxo2zqjrAT3SHzhr/zFta8uBNytm/R5JUlJWhxdOv09i//UOtBx4X4BbiaEzTVOJ3X2j9m3+X6Xb7yu1h4Rp28wPqeOIpAWwdAAAAACuk5WT5jf7YtC9RianJqu39mjbDprbRbdQqsq0iQ+JkV4xcrkhl5Hi0dF2+fizKkpQlaVu9fB/+bTEUFxWqVi2Kg47KAo/S8paRob51KSqTlJart75a5heISFIHV6r+cuh/vmDkz+P71igQkVQyLYxkkyHZj/nbbfbyiwp1IOOwkjPTdCD9sJIyUpWckaYDGYd1IPOwktJSlVWHKdsCpXYX1P2Dgsrv1q/pXfzlz8eoAKApm7vhF+1JTVaI3VE8OV5J0Ow3FNEonmbLZjP0e+pBfb/hF50+ZHTA2nw0lo0UKc/r9So1tfiXd+vWrfngqQJ3cVmvMDNdSx+6RRk7y/7wtYWEavS9Tyh+5AkBbBmq4yks0LrXn9XvP872K4/q0Fmj73taLbp0D1DLAAAAmhb6GKiJIrdHa3cdUmZekWIiQjSsR7saX2CvKdM0lZyZVrwA+r5dviAkKeNwrc9lyC6bomV6omUzW8gwW8gwo0vWoqgfhiHFRoapVYuSYCP6iMCjXHlsVKjsFl33yM1I12c3XKGYrAOVbt/vbK2tJ12j52/8g+XvGY6uKDtTPz9wk348tE8fx7epdt8h2TmKd0Ro4FnTFBoeUatRAX4X9hkVAKARcnm8WrPzYL3+LVGf6iUUQc3QYakfrrwcLXv0Th3etNZXZtjtGn7rQ+o0YUoAW4bK5B06oBVP3OMXZElSwujxGn7bQ3JGRgWoZQAAAE0PfQxUx+Xx6q25GzRz0ValZuX7ylu3CNefxvfV1ZMH1emChtfr1e+HD/mmv9qwd5c27ktURl527RtpOmSY0bKZMSXhRwsZZqQMWXOhpWVkqF+wUT7oaB0dVlIerrioMDka+OJO6QX3zF3Vj25p0b23Tnz8ZYVExzRQy5o20zRlejzyul3yut0yPW6/R6/HLbPCo0dej9vvmKLsLG3//H3lpiTr2U7tdTA0pNp62xUW6a69SYpsG6++F1ypkOgY2RwOGXaHbE6nbI7SL0fxo7PsuVG6zemQYbMTbACAxQhFAogOS/3xFBZo5dP3K/mXn8sKDUNDrr1L3aeeF7iGwc+htSv0y7MPqig7s6zQMNT/4mvVe9qlMhhlBgAAUCv0MVAVl8erG9/8SQs37pMpU17b75Ikm7ezDBVfcD1pYEe9/NdJlQYjLo9Xadn5Ss7I0Ybff9emfbu149Dv2peepNScQ3J767CAtRkiw2xRPPrDW/woRfjaU1MxESFlIzlKgo4jR3a0bhGmuOjwRnsXa00DkVIx3fvohMdfqtdgxPSUBgNHBgmeSoMEr9tV7nn5YzxVhhBel6uG5y7ZXlrmrqxdVQQcHmvXjFkbFaEP4tvWaN9Lkw9paM4xTrVlGBXCE8PuOCJMKbfd4ZThdFQRupQPXhwVjisNZozyx1YR4PjaUG67YSfAAYJJ6fW8phjSWxqKrFy5UrNmzdL27duVlZVV6ZyghmFo3rx5VlXZpNFhqV9et1trXnhMexd851fe/+Jr1fv/LucXdQCZpqntn7+vzR++Lnm9vnJndAuNvPMxtRs2JoCtAwAAaLroY6Aqr367Xi/OLh5N77EdkNuxTpLkcA+V3Zvg229c3/bqmdBSqdn5SsnI0f6MZKXkHFSuK02mkSXTyJYMb2VVVM8M84UfxUFIjKTQKgOQ6HBnpcGGX+DRIlytosIU4mzaC2XUNhApFdoyTgljJ8hmt1cfENQkQCgJJHzb3C7/ueIhSfJKNRolUqp0tEjjjOLqhy80cdQ1mDlitIxf8FJJgGOvanv5cKdiwEOAAxyb0t9dkuo9pK8Pli20ftNNN+nVV1/1BSGGYfiFIqWv+cBBQ7E5HL7pl3Z987mvfPOHr6soJ0sDr7yZf48B4MrL0ep/PqYDyxb4lcd076PR059SZLv2gWkYAAAA0EwVuT2auWirJMmUKY9th2+bx7ZDNm+8JLdMI1sLtydqwW9ZJQFIjmSU9OtrmjuYkqHIcuFHyRRYClFkmLNsiqpywYbf1FXRxeFHqNOyyxWNkun1Kj/1kDJ2bdOvb/5d+SkHa32Owow0JX7733poHaqyITJCqSFO2WsYGKWGOLUhMkJDcpvOwuzHqnjKsTqMHAuAqkfelBt1U9VomfIBji+YKR8EHRnMVD2yx+YsCXeOCHzKt4HrV2hMjgzzf37gpiYXjFjyV8Z7772nV155RUOHDtUzzzyj119/XV988YW2bdumHTt26MMPP9R//vMf3XffffrrX/9qRZVAjRg2mwZfc6ecUdHa9sm/feU7vpwpV262ht5wr2z25v3HdmOS9fsurXjiXuXs3+NX3vnkMzT0urtlDw0LUMsAAACA5ictu0BLtyXpq+U7fGuIeG3JMm05vn1MW46KnD9JRlHtKzAN2Y1oRTjj1CqyrdrHJKhbmw6KbxnjN7KjdXTxOh3hIcHV9/K6XMpLOaCcpH3KTd6v3AN7lXtgv3IO7FPewSR5XXX4mTdDht0um91RfPG55OJy6UVq38Vqu9233fdYeiHat6/d75iyY8udu/Qid7nzVXuMwymbw+4rn5ifrxOff1hZe3bV6HuL7txdwx9/VfbQMJlul7wuty808LrLRuWYpa9driO2u0rKyu9X7jyl05ZVOK+r3LnKtpnl6kDTCnCqHnlzRIBz5H5OZxXBTCUBjl8w47/dPxwiwAlmlY1uzNy1rckFI5ZMn3XiiSdq9erVSkxMVNu2bXXFFVfo/fffl6fcvI3//ve/ddVVV2nOnDmaMoXFriWGtje0376cqY3/esGvrP3YCRpx12OyO2s29BV1t3/JPK154XG588vukDEcDg356x3qetof+aUJAABgAfoYwa3Q5dbqnYe0dGuSlmxJ0pZ9aZIkU26ZRro8ttTitUTqMP2Vw+ZQQky8urftpP4duuq4Lj00onsPxUZFWP1tNCnuggLlJu9T7oGS0CO5OPTIPbBPeSnJftMF1wdbaJjsvgv9JRf2a3Kxv3woUGkIUS58cNiPOJ+zmnPbjzi2ZP8q2tMUpzCq6XRnDbHuS11VWHy+fKBSSZhSMZg5IsDxuGsezFS1n2+6tyOOIzxscooDyorhilFtMFMuwCn9nDlibRyjmpE2/gGOXUcGPhXaUBKospZt7Rzt868xf+4dyZJQJDY2VsOHD9ePP/4oSbryyiv13nvvye12+/1yGzx4sNq0acOaIiXosDS8xB++1tqXn/T7w7TN0FEaM/1pOcKD+4/5+uL1uLX5/df0238/9CsPi2ujUfc9qVZ9BwWoZQAAAM0PfYzg4vWa2p6UriVbkrRka5JW7zyoQpdHpjwyjXR5bWnyGodlGpll02DVQERImAZ37q4BHbtpQIduGtixm7q1SZDD3rTX7qiropysstDjwH7llIQfuQf2qSAt1ZpKDFvJ2iA1u2u9KV14ao6a04XBxs40TZlej290THUBjumuKpgpX370kTWVnueIkT6VBjisxdPkVBngVDolWvlgpuJImwr7lV8zp/x6NkcZseMf4JS1L9ABTnMIhMuzZOxqQUGB2rVr53sdHh4uScrMzFTLli195YMGDdK3335rRZVAnXQ99Sw5I6O16tkHfX9spqxbqZ8fvFHHP/R8o/8P29QUZqbrl2ceUMqvq/zKWw04TqPumaGw2FYBahkAAADQNB3MyNWSLUlauvWAlm1L0uHsgpIQJFNe47C8jsMyjYxahSCSJNMpu3uAWke206IZlynEETwBiGmaKsxIU07SXr9prnIP7FNO8j65srMsqcfmDFFkfHtFJnRSZHwHRbXvqMj4jopM6KiItgly5+c2qwtOzVlIdIxOePylSt8v3h9rGYZRfIHa7pDUuKfcLh/gmJ7KwxRf8HKUqdLKgpfKR+wcOR1aVQGO6XFXEgyVjMAhwJHp8cjj8chTWBDophyVYbMfMSVa1dOYVR/MlFvrpooRO0dOyeYpKtKm915RbtLeo7azqUylZUkokpCQoEOHDvlet29fvFDy1q1bNWbMGF95cnKyvPU8dBQ4mg7HT5TzoX9o+Yy75Skonlc3fdsmLb7vOo179EWFxbUOcAubh/Ttm7XiyXuVn+q/YGCPsy/SwMtvlM0RXPMJAwAAAHWRW+DSLzuStXRrkpZuOaAdyRky5S0XgqTJNNLrNCWWH8Mlw5AumzC8WQYipsejvNSDxaFHkv80V7nJ+319w2PlCI8oDj0SOigqoST0KAk/wlu1qfZO3+outJfignvjUdn7xfsT3PwDnMav/BRq1QY41Ww3KxmxU6OAp9xxZvljCXCqZHo98hR55CkqDHRTjqopBCOWTJ81depUbd68WYmJiZKkRYsWacKECTrnnHM0a9YsGYahhQsX6uSTT9aIESO0fPnyY62yWWBoe2ClbduopQ/fJldO2V0/kfEdNO6xlxQZ3yGALWv6dn//pX59/Tm/od/20DANu/l+dRw/OYAtAwAAaN7oYzR9Hq9Xm34/7JsSa/3uFBV5SkeCpMlrO1wSgniOfjJJ3du21+ge/TV/8xolZ6ZVu29kSEutfvwVhTqbxgW9I3lcRco7eMBvQfPi9T72Kfdgkky325J6QmJii0d6JBSP8ij9ikroqJAWLY95jYyqpijhgnvjVPp+SeL9AeqJ6fHIWzr65ojpz8wq1q+pNMCpczBz5H5VtcEt01uz38/BoDH/3rLkL52pU6fqu+++0/LlyzVmzBiNHz9eo0aN0ldffaX27durffv22rBhg0zT1B133GFFlcAxi+szUOOfel1L/nazbx7Y3OT9WnTPXzXu0RfVokuPALew6fEUFWr9G3/Xnrlf+ZVHJnTU6OlPK6ZrzwC1DAAAAGi89qZml0yJlaQV2w8oI69QppFdvB6ILU1eZ5pk1OyCfudW7TSmZ3+N7TVAY3sOULuYOM1eu1QfL/vxqMfmFmXoh42/6A/HjT3Wb6neuPPzlJu8Tzkl01vllhvtkZeSbNndxOGt2xaHHfFlgUdx+NFBzogoS+qoCiMQmpbS96v0OQDrGXa77Ha77CGhgW7KUdU4wPG4jwhcqh5pU2WAU/7RdeS2I9feIcApz5KRIhkZGVqyZIn69++vbt26SZIOHjyov/zlL/ruu+/k9XrVsmVLTZ8+XXfeeecxN7q54C6uxiE3eb+WPHiTcpP3+8qcUS10/MPPK67PwAC2rGnJO5SsFU/eq4wdW/zK40eeoOG3P6yQqOjANAwAACCI0MdoGrLyCrV8e7KWlgQhe1KzZBo5MktGgniNNMmo2YLb7Vu20pieA3R8r4Ea02uAOsT6Twfs8Xp1+jN36beD+2p0vl7tOuq7u5+VLUALupqmqaLsrOIRHkn7SgKQsvCjMKP60S41ZdjsimiXUBZ4+Ka56qDIdu1lDw382gWMQAAA1CfT6y0OZ1zlApMjRtoUZqVr/evP1Wg9kfIae5hvSShSnby8PGVmZqpdu3YB+6OqsaLD0ngUpKVqyd9uVtaenb4ye1i4xtz/jNoOHRXAljUNh9b/ol+eeUBFWRllhYahfn+6Wn3+74pq580FAACAdehjNE4uj1frd6do6dYkLdmyX7/uSZXHzJVpO1y8LogtTTKKanSuti1iNbZnf40pGQnSuVW7aqdr+nb9Ct36wYu+16Yk0+2WqeJLAYYMGQ6Hyp/hn5fcrNOHjK7Lt1ojpmmqIC21ZITHPuUklZvm6sB+uXKzLanHFhJaHHIkdFBUyeLmpVNdRbSJbxLrHBZlZ0piBAIAIHCqmtaxKo09EJEsCkVuv/12tWzZUn/729+saFPQoMPSuBRlZ2rpI7cpfdsmX5nN4dSIux5Th+MnBrBljZdpmvrtvx9q0/uvSt6yhR2dkdEaceejih9xfABbBwAAEHzoYzQOpmlq98FM/VxuSqy8omzfKBCv7bBk1Gyh1FZRLTS6R9l0WN3btq/zmhUNuVaF1+NWfspB5R7wH+mRm1w87ZVVC8U6I6NKwo7ixc0jEzoqqmTUR1hsa27QAgDAAjUNRppCICJZFIqEhITorLPO0ueff25Fm4IGHZbGx52fp+VP3KOUdSvLCm02DbtpurqccmbgGtYIufJyteaFx5S0dL5feUy3Xho9/WkWqwcAAAgA+hiBk5ZdoKXbknxTYh3IKB4BUjwS5LBkFNToPDERkRrdo3/JlFgD1KtdR0tmXTjaxYy6XMTwFBUq92CS/9oeB/Yr58Be5R06INNjzVzloS1ji0OPIxc3j++okBYxx7ywOQAAOLr6+FsiUCwZK5qQkMAfIWgWHOERGvu3v2vVsw8qadmC4kKvV2teeFyunBz1POeigLavscjem6gVT96j7L2JfuWdJpymoTfcJ0dY4OffBQAAAOpTocut1TsPaenW4hBk094D/iNBQvJqdJ6o0HCN6tHPNxKkb/susls8uqEmd3dm7tqmnx+4qcLFDFdejnIP7C83zdV+5R7Yq9wD+5V/+JA1C5sbhsJbtyuZ5qpscfPi4KODnBGRx14HAAA4JiHRMTrh8ZcabNRpfbJkpMi1116rL7/8Urt371Z4eLgV7QoK3MXVeHk9bq19+Un9/uNsv/I+F1yhfn++JqhDwKRlC7T6+Ufkzi/r5Bl2uwZddZu6nzEtqH82AAAAgUYfo/54vaa2J6VryZYkLdmapF92/q4Cd4q8tjSZtsMyjdwanSc8JFQju/fV2J4DNKbnAA3s2E0Ou73e2l3becBDY1upVf8hKjicopwD+1SUmW5JOwy7XRHt2lcY6RGV0FER7RJkDwm1pB4AAFC/jvzboqkFIpJFoUhKSopGjhypwYMH67XXXlOHDkybUxN0WBo30zS18Z0XtePLmX7l3c+YpsF/vSPo5qY1PR5t/vB1bf/8fb/ysLjWGnXPE2rVf0iAWgYAAIBS9DGsdTAjV0u2JGnp1gNasjVRqbkHShZHT5Npq9li4KEOp4Z366MxPftrbM8BGty5p0IaaIHv2gYix8oeEqrI9mULmkeVCz/C27SVzd74FzYHAABHV/o3hqQmF4hIFoUiV155pVJSUvTNN98oNDRUxx13nLp06VLpqBHDMPSvf/3rWKtsFuiwNH6maWr7p+9q84ev+5V3mnCaht3yoGwN1JkJtMLMDK167kEdKr/WiqRW/Ydo1D1PKCyudYBaBgAAgPLoYxyb3AKXftmRrKVbk7R48+/acWi3zJJ1QUwjS6rBoGiH3a7juvTS2J4DNLbXQB3XpadCnSH13nZPYUHx+h4lC5ln/b5b+xf/KHd+zUaw1JQzMro48GhffpqrDopK6KTQ2FaMHAcAIEgUZWdKUpMLRCSLQpHaLPpmGIY8Fi221tTRYWk6dn3zuda/8ZzffLnxI0/QqHtmyB7avNfPSN+xRSueuFf5Kcl+5T3O/D8NvPKWoAmGAAAAmgL6GLXj8Xq16ffDWro1SQs37dG6Pb/JpdRyIcjRu8t2w6YhXXpqTM8BGtuzv4Z366PwepgKyjRNFWVnKTe5ZEHzkvCj+HG/CtJSLK3PHhau+JEnqEWX7iULnHdSZEKHJnnhAwAAoDxLrmbOnz/fitMAjVb3M6bJGRmt1f98RGZJqJf8y89a+vBtGvPgs3JGRAW4hfUj8Yevtf61Z+V1FfnK7CGhOu6m6eo04bQAtgwAAACom32p2fp5S/FIkKXbNyvbdbB4OiwjQ7IfPQSxGYYGdOxWMhJkgEZ066uoMGvW1jQ9HuUfPqTcA/uVk7zPt8B56aM7z9pRH1VpinODAwAA1JQlochJJ51kxWmARq3ThClyRkZqxVPT5S0qlCSlblyjn6dfr+MfeUGhMbEBbqF1PK4i/frmP5T43Rd+5ZHxHTR6+lOK6dY7QC0DAAAAaicrr1DLtydr8aa9WrB5ow5k7S1eHN3IkAyvVIM1zvsmdNHxvQZoTK8BGt2jn1qER9a5PZ7CAuUmJ/mFHaUhSN6hAzLd7jqfuzx7SKgi4jsovE1bZezYVuMF0wlEAABAc8e8N0AtxI88QeMe/qeWPXaH3Pl5kqSMndu06J5rNO6xlxTRpl2AW3js8lMPasVT9yl92ya/8nYjjteIOx5RSFSLALUMAAAAODqXx6v1u1P08+Z9+mHDr9p+cJc8xmGZRnpxCFKDXnD3Nu01rs8gHd9rgEb36K/YyOga12+apoqyMiuEHrnJxVNdFaSlHsN35y8kJlaR8e2L1/YoWdy8eKqrjn7re9R0wXUCEQAAEAwsWVMEdcN8v01Xxo6tWvLQLSrKyvCVhbdup3GPv6zoDp0D17BjlLJhtVY+fX+Fu8j6XnSV+l74Fxm1WD8IAAAADS8Y+ximaWr3wUwt2rxP36/foPW/b1OhN1VeI00yaraeZce4djqxNATp2V9toltWX6fHo7zUg0es61HymLzfummubDZFtG7nCzv8H9vXahrfowUjBCIAACBYWBKKTJo0qeYVGobmzZt3rFU2C8HYYWlOsvcmasnfblZ+6kFfWUhMrMY98oJa9ugTwJbVnmma2vHlTG169xWZ3rKOozMySsNvf0QJo04IYOsAAABQU8HSx0jLLtDPW/fp2zUbtWLnJmUWJpeEIDWbeqpti1Ya33eQju81UGN7DlB8y7gK+7gLCpR3cL9yKixqvk95KcnWTXMVGlYcdFQIPTooom2CbA7rJnioKhghEAEAAMHEklDEVoO7xw3DkGmaMgxDHk/N7tZp7oKlw9Kc5R1K1pK/3aSc/b/7yhwRkRr7t7+r9YDjAtiymnPn52nNizO0/+cf/cpbdO2p0fc9paj2nQLUMgAAANRWoPsYRdmZkmT5xfVCl1urdhzUN2s2aPG2DUrK3FsSgrhqdHxsRIzG9Rmok/oO1pieA9Qxrk3JNFcZfmFHTrnwozD9sGXtL57mqnhaqyPDj9CWcb5prhrCkcEIgQgAAAg2loQie/bsqbTc6/Xq999/13fffacXXnhBt912m/7617+qS5cux1plsxDoDgusUZiRpiUP3ep3t5UtJFSj73tS8SPGBbBlR5e9/3eteOIeZf++y6+840lTdNyN98kRFh6glgEAAKAuAtnHKL3YLumYL7J7vaa27U/T7DUbNW/jeu1M2S23UiWjqEbHR4dFaXSPAZrYb6COi2unuIIC5ZVMbZV7YJ9yDyYp98A+3zqBx8qw2RXepvw0Vx0UGd9RUQkdFNGug5wRdV+YvT5Y+V4BAAA0NQ22psicOXN01lln6euvv9bUqVMbospGj1Ck+XDl5mjZY3fo8KZ1vjLDbteI2x9Wx/GTA9au6iQtX6jVzz/iN9+xYbdr4JU3q8eZFzTo3WoAAACwRqD6GFaMPjiYkav/rdqk79av1eak35TvOSQZhTU6NsIRqoGx7TSyRYz6uD2KPZyqvINJyjt0QKZFMxXYQ8PKQo8jRntEtIm3dJqrhlBfo3oAAAAauwZdaH348OEKDw/Xzz//3FBVNmqEIs2Lu6BAK5++TwdXLS0rNAwNve5udTv93MA17Aimx6MtM9/Stk//7Vce2jJOo+59oslM+wUAAICKGrKPcdVN98ntccvpcen4lA1q4crR0phoSdLxmdnKckZpaZtBctmdctgdevulJ/2Ozy1w6fv1W/W/1au0bs82ZRQmS0Z+jeoOMw31Mg31zstTt7TDii9y6eiTOh9daMtYRcZXtqh5hwaf5spKpe9VTVT2XqFh8D41DbxPAIJRc/vsa9BbWXr06KFvv/22IasEGowjLExj7n9Wq59/RPsWzS0uNE2te/VpFeVkqfe0ywLeiSrMytSqv/9Nh9Ys9yuP6ztIo+59UuGt2gSoZQAAAGhq3B63jKICjT28SS3duVobFaH/tmklSYryeDQ0J0cjUjZraVwfuUPC5PF6tWTLbn2+coVW7tqiQ9n7ZRplo5ZVzZ/KoV6vuucXqFd+gXrmF6h9YVGdQhDDZldE2/jKFzVv177RTXNlFbfHLbebtT0bO96npoH3CUAwam6ffQ0aimzevFl2u70hqwQalM3h0Ig7HpEzMlq7v53lK9/8/mty5WRrwOU3BiwYydi5TSueuEd5hw74lXc/Y5oG/eVW2ZzOgLQLAAAATZPT49LYw5sU686VV9Lc2Ja+bd/HttTgnDxFeLLVsiBR88O6qO8dV8mtmoUgIV6vuhUUqmdegXrm56tjYZFq2pO0h4X7hR2li5tHxHdoktNcAQAAwFoN8tfg4cOH9dBDD2nLli2aMmVKQ1QJBIxhs2nIdXfJGRWt7Z+96yv/7b8fqig7S8fdcK+MBg4Hf/9pjta+8pS8RWVzMttCQnXcDfeq8yTW+AEAAEDtFGVn6viUDWrpLg451kdF6GBoiG/7wdAQPd6lozKcpV3OlGrP5/Ca6lpQPAqkZ36BOhcUVttZDW0ZV2GkR2RCR0W2a9+kp7kCAABA/bMkFOnevXuV27Kzs5WWlibTNBUZGaknnnjCiiqBRs0wDA249DqFREVr479f8pXv+eFrufJyNOKOR2R3hlRzBmt4XS79+vbz2j1nll95RNsEjZ7+tFr26FPvbQAAAEDzUrqoektXjiRVGCVSqiwQqchumupcUKieJVNidSkolLPccpeG3S5bizhlyqGi8GgVhUX7Hl3hUfLaS0Y5eyXtL5T275K0S+OPH6nxx7eqUN8Lr/1b2bl5tfo+77rpaoWG+v/NvnHzdn0554danee4wf11xuSJFcr/9cGnOnCw+rDoSNdd+Se1iov1K9uzd78++OTLao/z1GKxeY/Ho8efe6XG+19ywTnq0qmDX9nhtHS99s7MGp9DkhLatdFfLvm/CuXfzJ2vtb9urtW5zpl6qgb27+1XVlhYpGdfeqtW54mOjNAt111RoXzR0pVatPSXWp3rlJOO15iRFddvfOaFN1Xkckmq3ft0pDXrN2nODwtqdczoEUN16oRxFcpf/deHSkvPrNW5brnmMkVHR/mVbd+5W59+MadW5xnQt5f++IfJFco/+OQL7dmbVKtzXXnx+Wof39avLCn5kN758LNanadLp/a65II/1uqYUkf+f/q/P05V7x7d/PbJzs7RC2+8V6vzxsXG6Pq/XFyh/IcFS7Ri1bpanWvqqRM0bMiACuW1+RyQpBCnU3ff8tcK5ct/WasfFy6t5IiqFX+Wj6pQHuyf5Ufq1b2LLjj3DxXKP/nvbP22a0+tzsVnec3U5LO8ph6484YKZY35s/xYfkc1RpaEIomJiVVuczqd6ty5syZOnKh7771XvXv3rnJfoLnpde7Fcka10NpXnpS8XklS0pKftDwvV6OnPy1HWHi91Z1/+JBWPjVdaVs3+JW3HTZGI+98VCHRMfVWNwAAAJqn0kAkc9c2X9mRo0QqYzNNdSwsUq+8fPXML1DXgkLZZCjXHq78kGj1/+M0v1Ef4W3a6bt5P+v7z7+WTEn5kvJdktJKvirXr3ePSst/27VHGZlZtfpevaa3QllWdra2bt9Zq/MktKt83b7de/Yp8fd9tTpXZRdccvPya92m6pimanW+3Lz8CmVFLlet21RQUFhpefLBlFqfK+vEMRXKvKa31udpGdOi0vKU1LRan2vY4IoXnSVp+47dKiis/HuvjYzMrFq3qVuXjpWW79q9V8mHaneR113JxbKcnLxatym2ZeU/8z17k2p9rsJKfq6FhYWW/n85miP/P+XkVLyg7/Z4at2m+LaVf64cSkmt9bmOHzWs0vLanicsNLTS8rT0zFqfi8/ymgkPq/xnvi8pudbn4rO8Zvgsbz4sCUW83oofMACKdZ18lpyRkfrlub/JdLslSYfWrtCSB2/S2If+oZCoyj+cj0XqxrVa+fR0FWb4dxj7XHCF+l10dYNP3wUAAIDmqapRIqUiPB5ddDBVPfILFGaayrWHak2LHvotOlKFNqdkGAoLDdX1V9zUYG0GAABAcGOFOaABdBh3spwRUVo+4255CgskSWlbN2jxfddp3KMvKiy24hD/ujBNUzu//o82vvOSTG9ZuuuIiNSI2x9WwujxltQDAACA4BQSHaMTHn/JN1rkaKNE8ux2uWyGwkxTWRFxShp6quKcoYorf06ns9Jj42Jj1LeKu4Wr0qZ1XKXlvbp3qfWUKzbDVqGsRXR0rdsUX8Xdxd26dFRYFXf5VqWyn1VkRPhR27Ttt50qNztZtQxD6tOr5t9jZETF0e8hTmetf05V3YUd365Nrc/VIjq6QpnNsNX6PNGREZWWt2kdV+tzxcVWPlK/d89uvrvGa/M+HallTItat6ltm9aVlnfv1kktqxixURVHJTfeRUVF1LpN7ePbVVrepVP7Wp1HkkIrGbkQGhpa6zbVpe5SR/5/ioqq+G/KYbdb9u+pbZvWtT5XVXfR1/Y8fJbXXF0/y4/UsX18leX5VYzYqAqf5TVTk8/yY9GYP8uP5XdUY2SYZnP6dpqWsWPHSpKWLVsW4JagoRzeukHLHr5NrtxsX1lkQkeNe/QlRcbX/Q8tSXIX5GvtS09o36K5fuXRnbtr9PSnFd2h8zGdHwAAAI1fQ/UxirIz9cFl5+qt+BZHnTqrXWGRLk4u1BXvfKTIlrHV7ov6cfn1d8ntrtlc4A6HXe+++mw9twiV4X1qGnifAASj5vbZVzEurYM5c+Zo0qRJmj9/fpX7/PTTT5o0aZLmzp1b5T5Ac9eq7yCd+NTrCi03MiT3wD4tuuevytqzq87nzUnaq4V3XVUhEOlw4ima8Ny/CEQAAABgqZDoGH0Z3/OogYgkHQwN0X/i+xOIAAAAoFGwJBT517/+pdWrV2vUqFFV7jN69GitWrVK77zzjhVVAk1WTNeeGv/0m4poVzYypCAtRYvuu0Zp2zdV2L8oO1NF2ZlVnu/Ayp+14PbLlZW4w1dm2Owa+JdbNPKux+UIr3yYIAAAAFBXHq9Xu5y5Nd4/3ZnMWpQAAABoFCwJRdasWaMhQ4YoMjKyyn0iIyM1dOhQrVy50ooqgSYtKqGjxj/9hqI7dfOVubKztOSBG5WyfpWvrCg7Uz8/cJN+fuCmCsGI6fVqy0dvavljd8iVm+MrD20Zq3GPv6Re5/xJhmHU/zcDAACAoDN3wy/KN1wyTMkwJbtpVvplmIYMU8o3XPp+wy+BbnbQctgdcjjsNfuys/RooPA+NQ28TwCCUXP77LNkTZHw8HCde+65+uijj6rd789//rO++OIL5eXVblGk5oo1RVCYlallj9ym9HIjRGwOp0bePUOtBw71LWApSTHd++iEx19SSHSMinKytOrvD+ngqqV+54vtM0Cj731S4a0rX5wOAAAAzVug+hhF2Zla/MCNytq1XZLk7NBdE596lSmzAAAA0OhYMlIkIiJChw4dOup+hw4dUkjI0eecBYJFaIsYjXvsJbUZPMJX5nW7tOLJe/TTLZf6AhFJyty1TT8/cJNSN67RgtsurxCIdDv9XJ345OsEIgAAAGhwIdExOvHxlxXTvY9iuvfR5GdfJxABAABAo2TJSJFJkyZp+fLl2rFjh9q3b1/pPklJSerRo4dGjhypRYsWHWuVzQIjRVDKU1SoX559UAeWLzz6zoYhlftva3OGaOj1d6vLKWfWYwsBAADQFAS6j1E65WtIdExA6gcAAACOxpKRIpdccokKCgp0zjnnaP/+/RW279+/X+ecc46Kiop0ySWXWFEl0KzYQ0I16t4n1GH8qUffuVwgEtE2QeOfeZNABAAAAI1CSHQMgQgAAAAaNUtWPbnssss0c+ZMzZs3Tz179tTJJ5+s3r17S5K2b9+uefPmqbCwUBMnTtRf/vIXK6oEmh13Xq5y9u2p8f6O8Agd/+iLiu7QuR5bBQAAAAAAAADNhyXTZ0lSQUGBbr31Vr3zzjtyu91+2xwOh6688ko9//zzCg8Pt6K6ZiHQQ9vReBRlZ/otql5T5RdfBwAAAOhjAAAAANWzZPosSQoLC9Prr7+u33//XTNnztRTTz2lp556SjNnztTevXv1+uuvBzQQufzyy2UYRpVf1157baXH7dy5U5dccokSEhIUFhamnj176t5771VOTk4DfwdoruoaiEhli6+Xzt0MAAAAAAAAAKiaJdNnlRcfH68LL7zQ6tNaZsqUKYqPj69QXnpHVXlr1qzRhAkTlJ2drWHDhmn8+PFasWKFnn76aX3zzTf6+eefFRPDHfoAAAAAAAAAADQFlocijd29996rCRMmHHU/j8ejiy66SNnZ2XryySd17733SpKKiop03nnnafbs2br77rv1xhtv1HOL0dyFRMfohMdfYvosAAAAAAAAAKhnlkyf9dJLL8lut+ubb76pcp9vvvlGdrtdr7/+uhVV1ruvvvpK27dv18CBA3XPPff4ykNCQvTmm2/K4XDonXfe0eHDhwPYSjQXpcFITPc+NT6GQAQAAAAAAAAAaseSUOS///2v4uPjNXXq1Cr3Of3009WuXTt9/vnnVlRZ7/73v/9Jks4//3wZhuG3LSEhQSeeeKLcbne1QRBQG7UJRghEAAAAAAAAAKD2LAlFtm3bpoEDB1YID/wqstk0aNAgbd261Yoq6+y///2vbr75Zl177bWaMWOGVq5cWel+69atkyQNHz680u3Dhg2TJK1fv75e2ongVJNghEAEAAAAAAAAAOrGkjVF0tLS1Lp166Pu17p1a6WmplpRZZ299NJLfq8feOABTZ06VR988IHi4uJ85Xv27JEkdezYsdLzlJaX7gdYpbo1RghEAAAAAAAAAKDuLAlFWrVqpR07dhx1vx07dqhly5ZWVFlrQ4cO1ejRozVp0iR16tRJKSkpWrBggaZPn645c+bozDPP1OLFi2WzFQ+eycnJkSRFRkZWer6oqChJUnZ29lHrHjt2bKXlGzduVL9+/Xx1AT6GXcfd95RWz7hL2YnF/7eiu/bUcfc9pSLDriL+zQAAgCam9O9nWIM+BgAAAIJdXfsYlkyfNWbMGK1atUpLly6tcp+lS5fql19+0ZgxY6yostZuvfVWXXfdderTp48iIiLUpUsXXXbZZVq5cqXi4uK0dOlSzZo1KyBtAyrjjGqh4fc/q+iuPRXdtaeG3/+snFEtAt0sAAAAAAAAAGiyLBkpcuONN+qLL77Q2WefrVdffVXTpk3zrS9imqZmzZqlG264QYZh6MYbb7SiSst06NBBV1xxhf7+979rzpw5Ov/88yUVp0zp6enKzc2t9LjSO6+io6OPWseyZcsqLS+9u4u75lClqCiNf+JVSWLKLAAAAPjQxwAAAADqxpKRIhMnTtR9992nw4cP68ILL1RsbKyGDRumYcOGKTY2VhdccIFSUlJ011136ZRTTrGiSkv17t1bkpSUlOQr69KliyRp3759lR5TWl66H1BfQqJjCEQAAAAAAAAAwAKWhCKSNGPGDM2cOVP9+vVTVlaW1q1bp3Xr1ikrK0v9+/fXzJkz9dRTT1lVnaXS09Ml+a8fMnToUEnS6tWrKz1mzZo1kqQhQ4bUb+MAAAAAAAAAAIAlLAtFJOnCCy/Uxo0blZSUpOXLl2v58uVKSkrShg0bdOGFF1pZlWVKp/eSpBEjRvjKzzzzTEnSZ599JtM0/Y45cOCAFi9eLIfDoalTpzZcYwEAAAAAAAAAQJ1ZGoqUio+P16hRozRq1CjFx8fXRxW1snbtWs2cOVNFRUV+5VlZWbr66qv1yy+/KDIyUldeeaVv21lnnaXevXtr48aNevrpp33lRUVFuuaaa+R2u3XllVeqdevWDfZ9AAAAAAAAAACAujPMI4dBNENffvml/vjHPyo2NlYjRoxQmzZtlJycrLVr1yo9PV1RUVH65JNPKoz6WL16tSZMmKCcnBwNHz5cPXv21PLly7Vnzx4NHDhQP//8s2Ji6r7WQ+kiiFUtkggAAAAAtUEfAwAAAKiew8qT7d+/X19//bW2b9+urKysCtNOSZJhGPrXv/5lZbVHNXjwYN18881auXKlNmzYoLS0NDmdTnXt2lWXXXaZbrnlFnXt2rXCccOHD9fatWv18MMP68cff9SGDRvUsWNH3X333XrwwQcVFRXVoN8HAAAAAAAAAACoO8tGivz973/X9OnT5Xa7fWWlpzYMw/faMAx5PB4rqmzyuIsLAAAAgJXoYwAAAADVs2RNkdmzZ+uuu+5SfHy83njjDZ166qmSpO+//14vv/yyxo4dK9M0ddddd+mnn36yokoAAAAAAAAAAIBasSQUefHFF+VwOPTTTz/pqquuUkJCgiTp1FNP1fXXX6+ff/5ZDz/8sJ5//nmFh4dbUSUAAAAAAAAAAECtWBKKrFmzRmPHjlWPHj0klU2XVd6DDz6ozp07a8aMGVZUCQAAAAAAAAAAUCuWhCI5OTnq0KGD73VoaKgkKTs721dmGIZGjhzJ3LYAAAAAAAAAACAgLAlF2rVrp7S0NL/XkrRjxw6//TIyMpSfn29FlQAAAAAAAAAAALViSSjSs2dP7d692/d6zJgxMk1Tr7/+uq9sy5Ytmj9/vnr27GlFlQAAAAAAAAAAALViSSgyZcoU/fbbb9q4caMk6ZRTTlGvXr309ttva9SoUTrvvPM0duxYuVwuXXPNNVZUCQAAAAAAAAAAUCsOK05y8cUXKzIyUqZpFp/U4dBXX32ladOmadWqVVq1apVsNpuuu+46XXvttVZUCQAAAAAAAAAAUCuWhCLt27fXDTfc4FfWt29fbdy4Udu2bVNaWpp69eql1q1bW1EdAAAAAAAAAABArVkSilSnT58+9V0FAAAAAAAAAADAUVmypggAAAAAAAAAAEBjRygCAAAAAAAAAACCAqEIAAAAAAAAAAAICoQiAAAAAAAAAAAgKBCKAAAAAAAAAACAoEAoAgAAAAAAAAAAggKhCAAAAAAAAAAACAqEIgAAAAAAAAAAICg46nLQpEmT6lyhYRiaN29enY8HAAAAAAAAAACoizqFIgsWLKi03DAMmaZZ7TbDMOpSJQAAAAAAAAAAwDGp0/RZu3fvrvB18803yzAMTZs2TV9++aXWrVundevW6csvv9T5558vm82mW265Rbt27bL6ewAAAAAAAAAAADiqOo0U6dKli9/rTz75RC+99JL++9//6uyzz/bbNnjwYJ111ln6+uuvde6552r06NEVjgcAAAAAAAAAAKhvhlnVfFe1MHr0aIWGhmrRokXV7jd+/HgVFBRo5cqVx1plszB27FhJ0rJlywLcEgAAAADNAX0MAAAAoHp1mj7rSJs2bVKnTp2Oul+nTp20efNmK6oEAAAAAAAAAACoFUtCkdDQUK1Zs+ao+61Zs0ahoaFWVAkAAAAAAAAAAFArloQikyZN0vbt23XnnXfK5XJV2O5yuXT33Xdr+/btmjRpkhVVAgAAAAAAAAAA1Iola4r89ttvGjVqlLKyshQfH68//vGP6tKliwzD0J49e/TFF1/owIEDatGihVasWKHevXtb0fYmj/l+AQAAAFiJPgYAAABQPYcVJ+nVq5cWLFigSy+9VBs2bNCrr74qwzAkSaWZy8CBA/X+++8TiAAAAAAAAAAAgICwJBSRpCFDhmj9+vVatGiRFi1apP3790uSOnTooBNPPFEnnXSSVVUBAAAAAAAAAADUmmWhSKnx48dr/PjxVp8WAAAAAAAAAADgmFiy0PqRCgsLdeDAAaWlpdXH6QEAAAAAAAAAAGrN0lDkrbfe0rBhwxQVFaWOHTvqzjvv9G2bNWuWzj33XO3YscPKKgEAAAAAAAAAAGrEklDE6/Xq/PPP17XXXqvNmzerb9++vgXWSw0ePFhffvmlPv30UyuqBAAAAAAAAAAAqBVLQpHXXntNs2bN0uTJk5WYmKgNGzZU2KdXr17q3r27vv32WyuqBAAAAAAAAAAAqBVLQpF///vfatOmjT799FPFx8dXud/AgQO1Z88eK6oEAAAAAAAAAACoFUtCka1bt2rMmDGKjo6udr+oqCgdOnTIiioBAAAAAAAAAABqxZJQxGazyeVyHXW/vXv3KjIy0ooqAQAAAAAAAAAAasWSUKRPnz5as2aN8vPzq9wnPT1d69at0+DBg62oEgAAAAAAAAAAoFYsCUX+7//+T4cOHdKtt94qr9db6T533nmncnJydNFFF1lRJQAAAAAAAAAAQK04rDjJTTfdpI8//lhvv/22li9frj/84Q+Sitcaeeyxx/TVV19pzZo1Gj16tK688korqgQAAAAAAAAAAKgVS0KRsLAw/fDDD7riiis0e/ZsbdiwQZK0fPlyLV++XJI0depUffDBB3I4LKkSAAAAAAAAAACgVixLKFq1aqWvv/5av/76q+bOnavdu3fL6/WqU6dOmjx5skaMGGFVVQAAAAAAAAAAALVm+bCNwYMHs5g6AAAAAAAAAABodCxZaB0AAAAAAAAAAKCxsyQUmTlzprp3767vv/++yn2+//57de/eXZ9++qkVVQIAAAAAAAAAANSKJaHIhx9+qOzsbE2aNKnKfSZOnKisrCx98MEHVlQJAAAAAAAAAABQK5aEIhs2bNDgwYPldDqr3CckJERDhgzRr7/+akWVAAAAAAAAAAAAtWJJKJKSkqKEhISj7hcfH69Dhw5ZUSUAAAAAAAAAAECtWBKKxMTE6Pfffz/qfvv27VNkZKQVVQIAAAAAAAAAANSKJaHI8OHDtXz5cm3durXKfbZu3aqlS5dq+PDhVlQJAAAAAAAAAABQK5aEIldffbXcbrfOPPNM/fLLLxW2r1q1Smeeeaa8Xq+uuuoqK6oEAAAAAAAAAACoFYcVJ/njH/+oP//5z/roo480ZswY9evXT71795Ykbd++XVu2bJFpmrrooot0/vnnW1ElAAAAAAAAAABArVgSikjS+++/rwEDBujZZ5/V5s2btXnzZt+22NhY3XXXXbrnnnusqg4AAAAAAAAAAKBWDNM0TStP6HK5tHr1at/C6507d9bw4cPldDqtrKZZGDt2rCRp2bJlAW4JAAAAgOaAPgYAAABQPctGipRyOp0aM2aMxowZY/WpAQAAAAAAAAAA6syShdYBAAAAAAAAAAAaO0tHiixevFgLFixQUlKSCgsLK93HMAz961//srJaAAAAAAAAAACAo7IkFMnJydG5556refPm6WhLlBCKAAAAAAAAAACAQLAkFHnggQf0448/qmvXrrr++uvVt29fRUdHW3FqAAAAAAAAAAAAS1gSisyaNUutWrXS8uXL1bZtWytOCQAAAAAAAAAAYClLFlpPSUnRCSecQCACAAAAAAAAAAAaLUtCkU6dOh11LREAAAAAAAAAAIBAsiQU+dOf/qSFCxcqIyPDitMBAAAAAAAAAABYzpJQ5P7779fQoUM1depUbdmyxYpTAgAAAAAAAAAAWMqShdZPO+00uVwuLV++XIMGDVLnzp3VuXNn2WwVMxfDMDRv3jwrqgUAAAAAAAAAAKgxS0KRBQsW+J57vV4lJiYqMTGx0n0Nw7CiSgAAAAAAAAAAgFqxJBTZvXu3FacBAAAAAAAAAACoN5aEIl26dLHiNAAAAAAAAAAAAPXGkoXWAQAAAAAAAAAAGjtCEQAAAAAAAAAAEBQsmT6r1KeffqpZs2Zp+/btysrKkmmaFfYxDEM7d+60stqjcrlcmj9/vmbPnq0FCxZo586d8ng86tixo6ZMmaK777670inAjrYo/LJlyzRmzJj6ajYAAAAAAAAAALCQJaGIx+PRueeeq9mzZ1cahEjFAUNV2+rbwoULNWXKFElSp06dNHnyZEnSypUr9eqrr+rDDz/Ud999p7Fjx1Y4NjIyUtOmTav0vG3atKm/RgMAAAAAAAAAAEtZEoq89NJL+t///qfTTz9dL7zwgh577DF9+OGHKigo0I4dO/Thhx/q+eef12233aYZM2ZYUWWt2Gw2TZs2Tbfffrtf8FFQUKDrrrtO7777ri666CL99ttvcjqdfse2bt1a7777bgO3GAAAAAAAAAAAWM2SNUVmzpypmJgYffzxx+rZs6dstuLTOp1O9evXTzNmzNB///tfPfXUU5o5c6YVVdbKpEmT9Nlnn1UYCRIWFqZXX31VMTEx2rNnj5YuXdrgbQMAAAAAAAAAAA3DklBk69atGj16tFq0aCGpbC0Oj8fj2+e0007TyJEj9fLLL1tRpWXCw8PVu3dvSVJSUlKAWwMAAAAAAAAAAOqLZWuKxMXF+V5HRERIkjIyMtSqVStfeY8ePTR79mwrqrSMx+NRYmKiJCk+Pr7C9tzcXD3xxBPas2ePwsLCNGDAAJ199tlq165dA7cUAAAAAAAAAAAcC0tCkfbt2+vAgQO+1126dJEk/frrr5o4caKvfPfu3XI4LKnSMh988IFSUlLUpk0bHX/88RW2p6am6v777/cru+WWW/TEE0/otttua6hmAgAAAAAAAACAY2RJQjF48GAtW7bM93rixIkyTVMPP/ywRowYoejoaH300Udavny5xo8fb0WVlkhMTNQdd9whSZoxY4ZCQ0P9tl9yySW66KKLNHjwYMXGxmrnzp1688039dprr+n2229XRESErrnmmqPWc+RaJqU2btyohIQEffnllzrllFN85YsXL1Zubq4mTJigsLAwSdKaNWt06NAhjRw50jf6Ztu2bdq9e7cGDBigTp06SZL27t2rTZs2qWvXrurbt68kKT09XStWrFCbNm00fPhwSVJhYaHmz5+viIgIv/dk3rx5crlcmjx5sm9tmOXLlysjI0Pjxo1TdHS0JGnDhg3av3+/jjvuON+omV27dmn79u3q3bu3unfvLkk6ePCg1q5dqw4dOmjQoEGSpOzsbC1ZskQtW7bUmDFjJEler1dz586V0+nUySef7GvPokWLlJeXp4kTJ/ren9WrVyslJUWjR49WbGyspOIp3BITEyv9WXTr1k19+vSRJB0+fFi//PKL2rZtq2HDhkmSCgoKtGDBAkVGRurEE0/01f3jjz/K7XbrtNNO85UtW7ZMmZmZOvHEExUZGSmpOPxLSkrSsGHD1LZtW0nSzp079dtvv6lPnz7q1q2bJCk5OVnr1q1Tx44dNXDgQL+fRWxsrEaPHi2pePTSDz/8oJCQEE2aNMlX98KFC5Wfn69JkyYpJCREkrRq1Sqlpqb6/Sy2bNmiPXv2aNCgQerQoYMkac+ePdqyZYu6d+/umy4uNTVVq1atUrt27XTcccdJkvLz87Vw4UJFRUXphBNO8NX9ww8/yOv1asqUKb6ypUuXKisry+9nsX79eh04cEDDhw9XmzZtJEk7duzQjh071LdvX3Xt2lWSdODAAa1fv16dOnXSgAEDJEmZmZlatmyZ4uLiNGrUKEmS2+3Wjz/+qNDQUL+AdcGCBSooKNDJJ58sp9MpSfrll190+PBhjR07VjExMZKkzZs36/fff9fgwYPVvn17v59Fjx491KtXL0lSSkqKVq9erfj4eA0dOlSSlJeXp0WLFik6Olrjxo3z1T137lxJ0uTJk31lS5YsUXZ2tsaPH+8bKbdu3TolJyf7/Sx+++037dy5U/369fOFx0lJSfr111/VuXNn9e/f3+9n0apVK40cOVKS5HK5NG/ePIWFhWnChAm+uufPn6/CwkKdcsopvtB55cqVSktL8/tZbNq0SXv37tWQIUOUkJAgqfgzcOvWrerZs6d69uzp97NISEjQkCFDJBWPmFu8eLFatGjhFx5///33stlsOvXUU31lP//8s3JycnTSSScpPDxckrR27VodPHhQI0aMUOvWrSVJ27dv165du/x+Fvv379eGDRvUpUsX9evXT1LZ51fr1q01YsQISVJRUZF++uknhYeH66STTvLV/dNPP6moqEinnnqq7Ha7JGnFihVKT0/3+/zauHGj9u3bp6FDh/pGCO7evVvbtm1Tr1691KNHD0nSoUOHtGbNGrVv316DBw/2+1nExMT4fbZ/9913cjgcfJbzWc5nOZ/lfJZX81k+bdo0wTr0Mfi9JPF7id9L/F6ij8FnOZ/lfJbzWR7cn+V17WNYsqbI1KlTlZycrHnz5kmSRowYoalTp2rx4sVq1aqVWrVqpUsvvVSGYeihhx6yospjlpWVpbPPPltpaWk6//zzdfXVV1fY5/3339fpp5+uDh06KCIiQoMGDdJLL73kWxdl+vTpKiwsbOimAwAAAAAAAACAOjBM0zSP9SQFBQXauXOn2rVr50t1cnNzdc8992jWrFlKS0tT37599cADD+j8888/5kYfq4KCAp122mlauHChTj75ZH3zzTcVRolUx+v1Kj4+XikpKVqwYIHfXQS1UXoHQPlRNgAAAABQV/QxAAAAgOpZMn1W6QLk5UVGRurll1/2japoLFwul8477zwtXLhQY8aM0VdffVWrQESSbDabevbsqZSUFCUlJdVTSwEAAAAAAAAAgJUsmT6rqfB6vbr44os1Z84cDRkyRHPmzPHNbVdb6enpklTn4wEAAAAAAAAAQMMKmlDENE1dffXV+vTTT9WnTx/NnTvXt9BQbW3cuFFbt26VJN9CVAAAAAAAAAAAoHGr0/RZkyZNqnOFhmH4FmRvSHfccYfeeecddevWTfPmzVPbtm2r3f+9997ToEGDNGzYML/yNWvW6E9/+pMk6bzzzlOHDh3qrc0AAAAAAAAAAMA6dQpFFixYUOcKDcOo87F19dVXX+n555+XJHXt2lX3339/pfudc845OueccyRJX3zxhS6//HL16dNH/fv3V0hIiHbs2KG1a9fK6/Vq+PDhevPNNxvqWwAAAAAAAAAAAMeoTqHI7t27rW5HvSpd/0OS5s+fX+V+Xbt29YUil156qSIiIrR27VotWLBA2dnZatGihcaPH6//+7//01/+8heFhITUd9MBAAAAAAAAAIBFDNM0zUA3IliNHTtWkrRs2bIAtwQAAABAc0AfAwAAAKhe0Cy0DgAAAAAAAAAAghuhCAAAAAAAAAAACAp1WlOkKp9++qlmzZql7du3KysrS5XNzGUYhnbu3GlltQAAAAAAAAAAAEdlSSji8Xh07rnnavbs2ZUGIVJxGMLyJQAAAAAAAAAAIFAsmT7rpZde0v/+9z+ddtpp2r59uy655BIZhqHCwkJt2rRJ9913n0JDQ3XffffJ6/VaUSUAAAAAAAAAAECtWDJSZObMmYqJidHHH3+sFi1ayGYrzlqcTqf69eunGTNm6MQTT9QZZ5yhAQMG6E9/+pMV1QIAAAAAAAAAANSYJSNFtm7dqtGjR6tFixaSiqfKkoqn1Sp12mmnaeTIkXr55ZetqBIAAAAAAAAAAKBWLAlFPB6P4uLifK8jIiIkSRkZGX779ejRQ5s2bbKiSgAAAAAAAAAAgFqxJBRp3769Dhw44HvdpUsXSdKvv/7qt9/u3bvlcFgyYxcAAAAAAAAAAECtWBKKDB48WNu2bfO9njhxokzT1MMPP6zs7GxJ0kcffaTly5dr0KBBVlQJAAAAAAAAAABQK5aEIlOnTlVycrLmzZsnSRoxYoSmTp2qxYsXq1WrVmrVqpUuvfRSGYahhx56yIoqAQAAAAAAAAAAasWSUOTPf/6zNmzYoCFDhvjKPvnkE11//fVq1aqVcnJyNHDgQP3nP//RxIkTragSAAAAAAAAAACgVgzTNM1ANyJYjR07VpK0bNmyALcEAAAAQHNAHwMAAAConiUjRQAAAAAAAAAAABo7R32ePCsrSzt27FBkZKT69OlTn1UBAAAAAAAAAABU65hCkfXr12vhwoVyu93q16+fTjvtNBmGIZfLpdtuu01vvvmmPB6PJKlXr1567733NHr0aEsaDgAAAAAAAAAAUBt1DkVuuukmvfrqq35lw4YN008//aTp06dX2LZ9+3addtpp2rhxozp06FDXagEAAAAAAAAAAOqkTmuK/Oc//9Err7wiSTrhhBN07rnnqlu3blqzZo3uv/9+vfnmmzrttNO0YcMG5ebm6tdff9XkyZOVmZmpf/zjH5Z+AwAAAAAAAAAAADVRp5Eib731lgzD0P/+9z9NnTpVkuTxeHTZZZfplVdeUcuWLfXpp58qKipKkjRw4EB99tln6tq1q+bOnWtd6wEAAAAAAAAAAGqoTiNF1q9fr5EjR/oCEUmy2+164IEHZJqmRo4c6QtESkVHR2vkyJFKTEw8pgYDAAAAAAAAAADURZ1CkfT0dPXo0aNCeWlZQkJCpcfFx8crLy+vLlUCAAAAAAAAAAAckzqFIqZpKjQ0tEK50+mUJBmGUelxVZUDAAAAAAAAAADUtzqFIgAAAAAAAAAAAE1NnRZal6Tk5GQtWrSoVtuSk5PrWh0AAAAAAAAAAMAxMUzTNGt7kM1mO6apsDweT52PbU7Gjh0rSVq2bFmAWwIAAACgOaCPAQAAAFSvTiNFOnfuzPogAAAAAAAAAACgSalTKJKYmGhxMwAAAAAAAAAAAOoXC60DAAAAAAAAAICgQCgCAAAAAAAAAACCAqEIAAAAAAAAAAAICoQiAAAAAAAAAAAgKBCKAAAAAAAAAACAoEAoAgAAAAAAAAAAggKhCAAAAAAAAAAACAqEIgAAAAAAAAAAICgQigAAAAAAAAAAgKBAKAIAAAAAAAAAAIICoQgAAAAAAAAAAAgKhCIAAAAAAAAAACAoEIoAAAAAAAAAAICgQCgCAAAAAAAAAACCAqEIAAAAAAAAAAAICoQiAAAAAAAAAAAgKBCKAAAAAAAAAACAoEAoAgAAAAAAAAAAggKhCAAAAAAAAAAACAqEIgAAAAAAAAAAICgQigAAAAAAAAAAgKBAKAIAAAAAAAAAAIICoQgAAAAAAAAAAAgKhCIAAAAAAAAAACAoEIoAAAAAAAAAAICgQCgCAAAAAAAAAACCAqEIAAAAAAAAAAAICoQiAAAAAAAAAAAgKBCKAAAAAAAAAACAoEAoAgAAAAAAAAAAggKhCAAAAAAAAAAACAqEIgAAAAAAAAAAICgQigAAAAAAAAAAgKBAKAIAAAAAAAAAAIICoQgAAAAAAAAAAAgKhCIAAAAAAAAAACAoEIoAAAAAAAAAAICgQCgCAAAAAAAAAACCAqEIAAAAAAAAAAAICoQiAAAAAAAAAAAgKBCKAAAAAAAAAACAoEAoAgAAAAAAAAAAggKhCAAAAAAAAAAACAqEIgAAAAAAAAAAICgQihxFUVGRnnzySQ0YMEDh4eFq06aNzj33XK1ZsybQTQMAAAAAAAAAALVAKFKNoqIiTZkyRdOnT1dqaqrOPPNM9e3bV1988YXGjBmj77//PtBNBAAAAAAAAAAANUQoUo2nn35aCxYs0MiRI/Xbb7/p008/1eLFizVz5ky5XC5dfPHFys7ODnQzAQAAAAAAAABADRCKVMHtduuf//ynJOnVV19VixYtfNsuuugiTZ06VampqXrnnXcC1EIAAAAAAAAAAFAbhCJVWLJkidLS0tStWzeNGDGiwvYLLrhAkvTVV181dNMAAAAAAAAAAEAdEIpUYd26dZKk4cOHV7p92LBhkqT169c3VJMAAAAAAAAAAMAxcAS6AY3Vnj17JEkdO3asdHtpeVpamnJychQVFVXlucaOHVtp+apVqxQaGqpRo0YdY2sBAACAxm3IkCF66623At2MZoM+BgAAAIJdXfsYhCJVyMnJkSRFRkZWur18CJKdnV1tKFIVh8OhyMhI2e32ujXyGG3cuFGSNHDgwIDUj5rhfWo6eK+aBt4nAMGIz77gQR8DNcV71TTwPjUNvE8AglFT/uwjFGkAy5YtC3QTKlV6d1ljbR+K8T41HbxXTQPvE4BgxGdf89NY30v+rTUdvFdNA+9T08D7BCAYNeXPPtYUqULpyI/c3NxKt5eOJJGk6OjoBmkTAAAAAAAAAACoO0KRKnTp0kWStG/fvkq3l5bHxcXVaeosAAAAAAAAAADQsAhFqjB06FBJ0urVqyvdvmbNGknFi7kAAAAAAAAAAIDGj1CkCuPGjVNcXJx2796tVatWVdj+ySefSJLOPvvshm4aAAAAAAAAAACoA0KRKjgcDt16662SpOuvv15ZWVm+bR9//LHmzJmj1q1b68orrwxQCwEAAAAAAAAAQG0YpmmagW5EY1VUVKQpU6ZowYIFatu2rU466SQlJydr8eLFcjqd+vrrr3XaaacFupkAAAAAAAAAAKAGCEWOoqioSM8995w+/PBD7dq1S1FRUTrhhBP0t7/9TcOGDQt08wAAAAAAAAAAQA0RigAAAAAAAAAAgKDAmiIAAAAAAAAAACAoEIoAAAAAAAAAAICgQCgCAAAAAAAAAACCAqEIAAAAAAAAAAAICoQiQaioqEhPPvmkBgwYoPDwcLVp00bnnnuu1qxZE+imoZxt27bpyiuvVNeuXRUaGqrIyEgNGTJEjzzyiHJycgLdvKCxZs0aPfPMMzr//PPVtWtXGYYhwzCUmJh41GNXrVqliy++WJ06dVJoaKjatm2rcePG6dlnn63/hgcZl8uluXPn6uabb9bgwYMVGRmpsLAw9ezZUzfccIP27NlTo/Ns27ZN4eHhMgxDY8aMqedWA4C1TNPUpEmTfL+rtm7dWul+n3zyiU4++WS1atVKTqdTbdq00ZQpU/Tll182bIPRrNDHaBroYzQe9DMaP/oYAFCsOfYzDNM0zUA3Ag2nqKhIU6ZM0YIFC9S2bVuddNJJOnDggH7++Wc5nU7973//05QpUwLdzKC3ePFiTZkyRfn5+erVq5eGDBmi3NxcLVmyRFlZWerXr5+WLFmi2NjYQDe12TvnnHP01VdfVSjfvXu3unbtWuVx//jHP3TXXXfJZrNpzJgx6tixow4dOqSNGzcqOjpaO3bsqMdWB58ff/xRp556qiSpU6dOGj58uCRp5cqVSkpKUosWLfTdd99p7NixVZ7D6/XqxBNP1LJly2SapkaPHq3ly5c3SPsBwApvvPGGrr32WhmGIdM0tWXLFvXt29dvnxtvvFGvvPKKbDabTjjhBCUkJCgxMVErVqyQJN1zzz166qmnAtF8NGH0MZoG+hiNC/2Mxo8+BgAUa5b9DBNB5dFHHzUlmSNHjjQzMzN95TNnzjQlma1btzazsrIC2EKYpmkOGDDAlGQ++OCDptfr9ZUfPnzYHDZsmCnJvPfeewPYwuDx1FNPmQ888ID55Zdfmvv27TPbtWtnSjJ3795d5TGffvqpKckcOnSouWPHDr9tbrfbXLlyZT23OvjMmzfPnDZtmrl06VK/8vz8fPPyyy83JZldunQxi4qKqjzHCy+8YEoyr732WlOSOXr06PpuNgBYZu/evWaLFi3M0047zezSpYspydyyZYvfPitXrjQlmVFRUeaaNWv8ts2dO9d0Op2mYRgVjgOOhj5G00Afo3Ghn9H40ccAgObbzyAUCSIul8uMi4szJZm//PJLhe1Tp041JZn//Oc/A9A6lEpNTTUlmU6n0ywsLKywvbRzOXHixAC0DkfrrBQUFJjt2rUzIyMjzb179zZs41CpvLw8MyYmxpRkLliwoNJ9du3aZUZGRppnnHGGOX/+fDosAJqcqVOnmpGRkWZiYmKVnZXnnnvOlGRedtlllZ5j8uTJpiTzvffea4AWo7mgj9E00Mdo/OhnNC30MQAEi+baz2BNkSCyZMkSpaWlqVu3bhoxYkSF7RdccIEkVTqEFw0nJCSkRvu1bt26nluCuvjiiy908OBBnX/++erYsWOgmwNJ4eHh6t27tyQpKSmp0n2uvvpqGYah1157rSGbBgCW+OCDDzRnzhw9/vjj6tKlS5X7hYaG1uh8/I2B2qCP0TTQx2j66Gc0LvQxAASD5tzPIBQJIuvWrZMk3zyYRxo2bJgkaf369Q3VJFQiOjpaxx9/vFwulx5//HGZ5Zb9SUtL03PPPSdJuuqqqwLVRFRj3rx5kqQTTjhB2dnZeuutt3TDDTfolltu0bvvvqvc3NwAtzD4eDwe34KV8fHxFba//fbbmjdvnp544gl16tSpgVsHAMfm4MGDuvXWWzVy5EjdfPPN1e578skny263a9asWVq7dq3fth9//FHz589Xt27dNGnSpPpsMpoZ+hhNA32Mpo9+RuNCHwNAc9fc+xmOQDcADWfPnj2SVOVdJaXlaWlpysnJUVRUVIO1Df7efvttnXbaaXrsscf0n//8R0OGDFFeXp5+/vlnRUZG6oMPPtDkyZMD3UxUYtOmTZKK/x/1799f+/bt89t+33336YsvvtCYMWMC0byg9MEHHyglJUVt2rTR8ccf77dt//79uvPOOzV69GjdcMMNAWohANTdDTfcoKysLL311luy2aq/36lfv376xz/+odtuu00jR47UCSecoPj4eO3Zs0crVqzQiSeeqH//+98KCwtroNajOaCP0XTQx2ja6Gc0LvQxADR3zb2fwUiRIJKTkyNJioyMrHR7+Q5KdnZ2g7QJlevXr5+WLFmiMWPG6LffftPnn3+uOXPmKCsrS2PHjlX//v0D3URUIS0tTZJ0//33KzQ0VD/88IOysrK0ceNGnXPOOUpOTtYf/vAHHTx4MMAtDQ6JiYm64447JEkzZsyoMKTz2muvVV5eXo1+yQNAYzNr1izNmjVLd955p4YMGVKjY26++WZ99tlnioyM1MKFC/XJJ59o+fLliomJ0fjx49WmTZt6bjWaG/oYTQd9jKaNfkbjQR8DQHMXDP0MPp2BRmj+/PkaPHiwMjMz9d133yk9PV379+/XG2+8oXnz5mncuHGaO3duoJuJSni9XkmSaZr67rvvdMoppyg6OloDBgzQrFmzNGTIEB0+fFivvPJKgFva/GVlZenss89WWlqazj//fF199dV+22fOnKnZs2fr7rvv1qBBgwLUSgCom7S0NN1www3q2bOnHnrooRodY5qmbr/9dp133nk699xztWnTJuXm5mrTpk0666yz9Pjjj/umZQHQ/NDHaNroZzQO9DEANHfB0s8gFAkipXdpVTXXaOldXlLxnLMIjLS0NE2bNk2FhYX69ttvNWXKFLVs2VLt27fXX//6V73xxhsqKCjQtddeK4/HE+jm4gil/88mTpyonj17+m2z2Wy+P5oXLFjQ0E0LKgUFBTrrrLP066+/6uSTT9YHH3zgtz0lJUW33HKLevfurQceeCBArQSAurvtttt08OBBvfHGGzUehv7ee+/p+eef1xlnnKF///vf6t+/vyIiItS/f3+99957Ov300/Xrr7/61hYAaoI+RtNAH6Ppo58RePQxAASDYOlnsKZIEOnSpYskVZh7tFRpeVxcHHP9BtA333yjtLQ0nXzyyb73rLzzzjtPISEh2r17t3bt2qVevXoFoJWoSteuXbV27Vp169at0u2l5cnJyQ3ZrKDicrl03nnnaeHChRozZoy++uqrCkPalyxZotTUVEVFRem0007z25aRkSFJ2rx5syZMmCBJmj17Np+LABqVr776SmFhYXr00Uf16KOP+m0r/R1z6aWXKiIiQjfeeKOmTZvmu3hz4YUXVnrOCy+8UN9++61+/PFHPfLII/X7DaDZoI/RNNDHaProZwQWfQwAwSJY+hmEIkFk6NChkqTVq1dXun3NmjWSVOO54lA/SjuOMTExlW53OByKjIxUUVGRb15ZNB7Dhg3TF198UeV7c/jwYUnij9964vV6dfHFF2vOnDkaMmSI5syZU+Uc51LxfMCJiYmVbsvOztbChQslSW63uz6aCwDHpKCgwPc5VZlffvlFknTOOedIOvrfGC1btpQk/r5ArdDHaBroYzR99DMChz4GgGATDP0Mps8KIuPGjVNcXJx2796tVatWVdj+ySefSJLOPvvshm4ayklISJBU3IGs7I+k3377Tenp6ZKK7xZC41L6/2fp0qUqKCiosH3evHmSpOHDhzdou4KBaZq6+uqr9emnn6pPnz6aO3euYmNjK933nHPOkWmalX7Nnz9fkjR69GhfWekvcABoLDIyMqr8HCu9C3zLli0yTVO33nqrpLK/MVasWFHpOUvL+fsCtUEfo2mgj9H00c8IDPoYAIJNsPQzCEWCiMPh8P1jvf7665WVleXb9vHHH2vOnDlq3bq1rrzyygC1EJJ0+umnKzw8XImJibrnnnv8Oi2pqam66qqrJEknnXSS2rVrF6hmogqDBg3SGWecoaSkJN1xxx1+czJ/9dVX+vDDD2Wz2XTttdcGsJXN0x133KF33nlH3bp107x589S2bdtANwkAGpVzzz1XkvT8889r8eLFftsWLFig559/XpJ0wQUXNHjb0HTRx2ga6GM0ffQzAoM+BgAcXVPsZximaZqBbgQaTlFRkaZMmaIFCxaobdu2Oumkk5ScnKzFixfL6XTq66+/rjD3JRre22+/rWuuuUZer1edO3fWsGHDlJeXpxUrVigzM1Pt2rXTokWL1Lt370A3tdn75ptv9Nhjj/ler1mzRi6XS0OHDvXNIXvGGWfowQcf9O2TnJyscePGadeuXerWrZuOO+447d271ze88LnnntMdd9zRsN9IM/fVV1/5hm1OnDhRnTt3rnS/c845x7dfVRYsWKCJEydq9OjRWr58ucUtBYD617VrV+3Zs0dbtmxR3759feVFRUU644wz9OOPP8owDI0aNUpdunRRYmKiVq5cKal4XYFPP/1UNhv3TqHm6GM0DfQxGhf6GY0ffQwA8Nec+hmsKRJkQkJC9P333+u5557Thx9+qK+//lpRUVE6++yz9be//U3Dhg0LdBMh6aqrrtKgQYP0z3/+U0uWLNE333wjh8Ohbt266eqrr9Zdd93FHSoNJCUlpdLhf+vWrfM9L/+LQJLi4+O1evVqzZgxQ1988YVmz56tyMhInXbaabrjjjt0yimn1Hezg07pdA+SfEPTK9O1a9ejdlgAoLkKCQnRd999p7feekszZ87Uhg0btGrVKrVo0UITJkzQZZddpssuu0yGYQS6qWhi6GM0DfQxGhf6GY0ffQwAqJmm2M9gpAgAAAAAAAAAAAgKjWO8CgAAAAAAAAAAQD0jFAEAAAAAAAAAAEGBUAQAAAAAAAAAAAQFQhEAAAAAAAAAABAUCEUAAAAAAAAAAEBQIBQBAAAAAAAAAABBgVAEAAAAAAAAAAAEBUIRAAAAAAAAAAAQFAhFAAAAAAAAAABAUCAUAQAAAAAAAAAAQYFQBAAAAAAAAAAABAVCEQANorCwUC+//LImTpyotm3bKjQ0VPHx8Tr99NP17rvvyuPx1PncCxYskGEYuvzyyy1pa9euXWUYhiXnqgurvx+rTJgwQYZhKDEx0a+8qp+XYRjq2rVrwzTOAk2tvXVx+eWXyzAMLViwICDnaiw/48TERBmGoQkTJgS6KQAA4BjQx6g5+hiB0dTaWxf0MYrRxwCaFkIRAPVuy5YtGjBggG666SYtW7ZMgwcP1nnnnafevXvrxx9/1BVXXKFRo0bpwIEDgW4qmql3331XhmHo4YcfDnRT6izQHWlYq6oLAACA/2fvzuNqyv8/gL9u+66iPS122fcUlWixpGzZBlnGzliHmWHsxpjF7msd+zIkJDujkJ00lkiIEJVo06I6vz/63Ttd9942UfR6Ph49Zpz1c8/5nM+95/P+LERFw3cMKmt8x6Dyhu8YREWnUtYJIKKv2/Pnz+Hk5ISEhAR0794da9euRZUqVSTrnz59igEDBuDs2bNwdnbGjRs3oKOjU6xztGzZEhEREahUqVKppPn06dN4//59qRyrIvharldERARUVVXLOhmf1C+//ILp06fDysqqrJNCREREVGJ8x/j6fS3Xi+8YRETlE4MiRPRJjRw5EgkJCWjfvj327NkDZWVlqfVWVlY4evQoWrRogbt37+LHH3/E8uXLi3UOLS0t1KlTp9TSXL169VI7VkXwtVyv0sxD5ZWZmRnMzMzKOhlEREREH4XvGF+/r+V68R2DiKh84vBZRPTJREREICgoCACwbNkymZcVMS0tLfz6668AgA0bNuDNmzeSdfnH5UxNTcW0adNQo0YNqKurw8fHB0DB4+Omp6fj559/RvXq1aGurg5ra2tMnToVaWlpJR6/Njc3F0uXLkW9evWgoaEBExMT+Pn54dWrVzL7vHnzBitXrkTHjh1ha2sLDQ0N6Ovro02bNti8eXMRrmLR5eTkYNOmTXB1dYWhoSE0NDRgY2OD7t274/Dhw1Lbnj17FuPHj0fjxo1RpUoVqKurw9bWFsOHDy92V9vCulxnZ2dj4cKFqF27NjQ0NGBubo4RI0YgLi5OZtv8XdAfPnyIb775BmZmZlBWVsbSpUsB5LUMXLx4MVxdXWFlZQV1dXVUrlwZbm5uOHTokMwxXVxcMHjwYADAnDlzIBKJJH/5u7oXNBbtmTNn4OXlBSMjI6irq8PKygpDhw7Fo0ePZLbNn2czMzMxa9YsSZ61tLTEd999h5SUlAKuqDRx/n7y5Ikknfn/xGbPng2RSITNmzfjxo0b8PHxgbGxMZSUlHDgwAEAisfoDQsLw7Rp09CiRQuYmJhATU0NlpaW6NevH27dulXktBZVcfJE/s8lT0HjDkdGRqJXr14wNDSEtrY2WrZsid27dxeYtpycHPz555+ws7OTpG348OGIj48v8Fzp6en47bff0Lx5c+jq6kJbWxvNmjXDypUrpcYzF+ePkJAQAICtra3U/WRXdyIiooLxHYPvGADfMfiOIYvvGHzHICoO9hQhok/myJEjAIBGjRqhXr16BW7bsWNHGBgY4M2bNzhz5gy6d+8utT49PR3Ozs6IioqCs7MzmjRpgsqVKxd4zPfv38PT0xNnz56Fnp4eOnbsCCUlJaxfvx7nzp2DklLJ4sIDBgxAQEAAnJ2dUadOHVy8eBFbtmzB1atXcePGDairq0u2DQ0Nxbhx41C1alXUrFkT9vb2ePXqFS5cuIDQ0FBcvXoVq1atKlE68ktPT0fXrl1x6tQpqKurw9HREcbGxoiJicHJkyeRmJiIzp07S7afNGkSbt26hYYNG6Jt27YQiUS4ffs21q9fj3379uHChQuoXbv2R6dLEAT06tULR48eRbt27dC4cWOcO3cO69atw/Hjx3HhwgWYm5vL7BcZGSn54efk5IS0tDRoaWkBAPbv3y95ca1Tpw4cHBzw/PlznDlzBqdOncLixYsxdepUybE8PT2RnZ2N0NBQNGrUCI0bN5asy///iixZsgSTJk2CSCSCo6MjLC0tER4ejr/++gv+/v44evQoHBwcZPbLysqCh4cHwsLC4OzsjHr16uH8+fNYvnw57ty5g5MnTxZp/F5TU1MMGjQI/v7+SEtLw6BBgwrc/vz58xgxYgRsbGzQvn17JCQkFNplf8GCBThw4AAaNGiAli1bQkNDA/fv38euXbtw4MABHDt2DE5OToWmtShKmieKKzw8HM7OzkhKSkKtWrXQtGlTxMTEoG/fvhg/frzC/QYMGIBdu3ZBQ0MDrq6u0NHRQWBgIE6fPo369evL3SchIQHu7u4ICwuDkZERHB0doaqqikuXLmHcuHE4c+YM/P39IRKJoKOjg0GDBuHYsWN49eoVevToITWcR3GH9iAiIqpo+I7Bdwy+Y/Ad40N8x+A7BlGxCUREn0j//v0FAMLQoUOLtH27du0EAMLMmTMlyx4/fiwAEAAIzZo1E+Lj42X2O3PmjABAGDRokNTyRYsWCQCEevXqCa9evZIsj4uLExo0aCA57uPHj6X2s7a2FuQVj+Ltq1evLkRHR0uWJyUlCS1atBAACJs3b5baJyoqSggNDZU5VmxsrNC4cWMBgHDx4sUifZ6CjBgxQgAgNG3aVHj69KnUuuTkZOHUqVNSy4KCgoTExESpZbm5ucKaNWsEAIKHh4fMOZydnUt0vUxMTISIiAjJ8vT0dMHHx0cAIHTr1k1qn02bNkn2GzZsmJCVlSVz3Js3bwrh4eEyyyMjI4WqVasKKioqMtdAfNxZs2bJ7Jc/vdbW1lLLrl+/LigrKwvq6urCiRMnJMtzc3OFmTNnCgAES0tLIT09XbIuf55t2bKlkJCQIFkXGxsr2NjYCACE4OBghWmRR9F1Fps1a5bkvHPmzBFyc3Nlthk0aJAAQDhz5ozU8lOnTgkvXryQ2T4wMFBQVVUV6tSpI3M8RccqSEnyhPhzbdq0Se4x5aUjNzdXaNiwoQBAmDBhgpCTkyNZFxAQICgpKQkABGdnZ6lj7d69WwAgmJmZCZGRkZLlqampQocOHSTp//Aze3l5Scq61NRUyfK3b98KHh4eAgBhzZo1Uvsoep6IiIioYHzH4DsG3zH4jpEf3zH+w3cMoqLj8FlE9Mm8fv0aAGBkZFSk7Y2NjQHktYiQZ+XKlVITKBZm9erVAIBff/1Vcmxxen7//fciH+dDy5cvh7W1teTfenp6klZDH3Z5rV69utwWPqampli8eDEAYN++fSVOCwDExsZi48aNUFNTw/79+1G1alWp9bq6umjfvr3Uss6dO8PAwEBqmUgkwogRI+Dg4ICTJ08Wq/t1QX7++WepsXQ1NDSwatUqqKur48CBA5Iu2/lVrlwZS5Yskdv6qFGjRmjYsKHM8po1a2LmzJnIzs5GYGBgqaR9xYoVyMnJwbBhw+Dm5iZZLu4WX6dOHTx79gx79+6V2VckEmHjxo1SrQ1NTU0xevRoALJ5pbTUrVsXM2bMKFILMbH27dvLHQfYy8sLvXr1wr1793D37t1SS2NJ8kRxnD17Fv/++y9MTU2xaNEiqRab3bp1k2klKiZuUfnTTz+hZs2akuXa2tpYsWKF3GsaHh6OQ4cOoV69elizZg20tbUl6ypVqoRNmzZBTU0N//vf/z7qMxEREVEevmPwHQPgOwbfMWTxHYOIioPDZxHRF8HExAT29vZF3v7p06d4+vQpdHR00KlTJ5n17u7uMDQ0RGJiYrHSoaKiAnd3d5nl4h9fL168kFknCALOnj2Lc+fO4cWLF0hPT4cgCJIXgsjIyGKl4UPBwcHIzs5Gx44dYWVlVeT94uLicOjQIdy9exdJSUnIzs4GALx8+RK5ubmIiopCkyZNPiptANC/f3+ZZebm5mjXrh2OHTuG8+fPS70AAkCHDh0K7OL7/v17nDp1CpcuXcKrV6+QmZkJQRAQGxsL4OOvqdjZs2cBAN98843MOiUlJQwYMAA//fQTQkJCMGDAAKn1VlZWcrtCF5RXSkPXrl1LNGxDUlISgoKCEB4ejjdv3uD9+/cAgNu3bwPIu6aFDVFRVCXJE8Uhvm/dunWTGmpC7JtvvoG/v7/Usvfv3+Py5csAgN69e8vsU6dOHTRp0gQ3btyQWn7ixAkAeS93KiqyP6vMzMxQs2ZN3Lp1C+np6dDU1CzZhyIiIqJSx3cMxfiOwXeM/PiOwXcMoq8NgyJE9MmIW6/Ex8cXaXvxBGjyWmoV98eL+Mdg1apVFbZmsbKyKvYLi5mZmdwfJbq6ugCAzMxMqeUvX76Ej4+P5IeQPMnJycVKw4eePn0KAKhVq1aR91m9ejUmT56MjIyMT5YuANDX10elSpXkrhNPOPjs2TOZdQXd73v37sHb27vAl5LSSDuQN+EikDdRnTzVqlWT2i4/RS+P8vLKgQMHJBMV5vf7778Xq+UiUPxnBcgbQ3nIkCF4+/atwm1K65qWNE8Uh/h+KLoW8ia7fP36NbKysqCpqanwmltZWcm8sDx+/BgAsGjRIixatKjAdCUmJsLCwqKw5BMREVEB+I7Bdwy+Y8jiOwbfMYioeBgUIaJPpkmTJtixYweuXbtW6La5ubkIDw8HIH9iuvLS8qG4rWOGDRuGy5cvo2vXrvj+++9Rt25dVKpUCcrKyoiMjETt2rUhCMInSq18V69exdixY6Gjo4MVK1bA1dUVZmZmkmvcr18/7Nq167OnK7+C7nfPnj0RGRmJoUOHYtSoUahRowZ0dXWhpKSEEydOwMPDo0zTLlacvHLz5k1s2bJFZvns2bOL/cJS3GclJiYG/fr1Q05ODn777Td4eXnB0tISWlpaEIlE+PHHH/HLL7+Ui2sqT25ursJ1xene/7Hnb9WqlVR3fXnktSgjIiKi4uE7Bt8xSorvGHn4jlE4vmMQff0YFCGiT6ZTp06YMmUKwsPDcffuXdjZ2Snc9tixY0hMTISmpibatWv30ec2NzcHkPdjTBFx66dPJS0tDUePHoWxsTECAgKgrKwstT4qKqpUziNuLVTU7tz79u2DIAhYuHAhhg0bJrO+tNIFAG/fvkVycjL09PRk1kVHRwNAsVq13Lt3D3fu3EGzZs2wYcMGmfWlmXYgL22PHj3C48ePYWJiIrNe3ILnY1vmzJ49G7Nnz/6oY5TU4cOHkZGRgcmTJ2PKlCky60v7mhY3T6ipqQGAwvGn5T3H4v0VjRssPk9+lStXhpqaGtLT05GQkCD3RVHeucTja3fs2BGzZs2Sez4iIiIqPXzH4DsG3zGKhu8YefiOQUTycKJ1Ivpk6tati86dOwMAJkyYoLC1RXp6Or7//nsAea2ePpycrySsrKxgZWWF1NRUHD16VGb9qVOnit2tvbiSkpKQm5sLc3NzmZcVANixY0epnKddu3ZQUVHB6dOnC3xBExN/7g8nSwSAiIgIhIWFlUq6xHbu3Cmz7OXLlzhz5gxEIhHatGlT5GMVlHZF5wL++9ErHtO4qJycnAAA27dvl1mXm5uLbdu2AQCcnZ2LddySKOlnKExB1zQ+Ph4nT54s1fMBxcsT4sqH+/fvy03fh13NAaBt27YA8rrsfzjchKLzq6qqolWrVgAgd1LLyMhI3Lx5U2a5ePzvgwcPFqul26e6n0RERF87vmPwHQPgO0Zp4TsG3zGIKioGRYjok1qzZg2qVKmCkydPonfv3nj9+rXU+piYGHTq1Al37txBzZo1sXDhwlI79+jRowEA06ZNkxpz+PXr13Jbq5Q2ExMT6Ovr49atWwgODpZat2nTJuzatatUzmNqaoohQ4YgKysL3bt3lxl7NjU1FadPn5b8W9z9dt26dcjKypIsj4uLw6BBg0r9B9TcuXOlWphlZGRg7NixyMzMhJeXl9yxVxWpWbMmlJSUcPr0ady9e1eyPDc3F3PnzkVoaKjc/cQ/eiMiIoqV9rFjx0JJSQkbNmyQuoaCIGDevHmIiIiAhYUFevXqVazjlkRJP0NhxPlhy5YtUi2lUlJSCh0DuKSKkydcXFwgEomwbds2qX2SkpIwdOhQua27nJ2dUb9+fcTGxuKnn36SepEIDAzEvn375KZrzJgxAID58+fj4cOHkuXv3r3DuHHj5Fa6tGjRAp06dUJYWBiGDh0qtyIkMjJSZtLFT3U/iYiIKgK+Y/Adg+8YpYPvGHzHIKqoOHwWEX1SlpaWCAkJgZeXF/z9/REUFIQ2bdrAyMgIL168QGhoKLKzs9G0aVMcOnQIOjo6pXbuSZMm4ciRIzh79ixq1qyJdu3aQSQS4cyZM6hZsybs7e1x6dIlSWuK0qasrIwff/wR33//Pdq3bw9nZ2eYmpri1q1buH37NqZPn17opGlFtWTJEkRGRiI4OBjVq1dHmzZtYGxsjJiYGNy8eRPNmjVD+/btAQCDBw/GkiVLcOTIEVSvXh329vZIT09HSEgILCws4OPjI3dCvpKwsrJCkyZN0LBhQ7i6ukJXVxfnz5/HixcvYGVlhVWrVhXreEZGRhg5ciRWr16NJk2aoF27dtDX18fVq1fx5MkTTJkyBb///rvMfq1bt4aJiQkCAgLg5OSE6tWrQ1lZGV27dkXXrl0Vnq9Zs2b4/fffMWnSJLi5uaFt27awsLDAzZs3ERERAT09PezZswcaGhrFvjbF5ePjg5CQELRv3x6urq6SZ0VeF//i8PLyQqNGjRAWFoZq1aqhbdu2EAQBZ8+ehYqKCgYPHoxNmzaVxkcAUPw8YWNjg6FDh2LDhg1o2rSppGXdlStXYGRkBG9vbxw8eFBqH5FIhK1bt8LFxQV//PEHgoKC0LRpU8TExCA0NBTjxo3D8uXLZdLWu3dvHDx4ELt27UKDBg3g6uoKbW1tnD17FhoaGvDy8sKhQ4dkyoxt27ahY8eO2LRpE/z9/dG4cWNUrVoVqampuH37Nh49egRvb2/07NlTso+Pjw+2bNmC/v37w93dHfr6+gCAX3/9VTKBLBEREcnHdwy+Y/Ado3TwHYPvGEQVlkBE9Bmkp6cLy5cvF5ydnYXKlSsLqqqqgpGRkeDh4SFs2rRJyM7Olrvf48ePBQCCs7OzwmOfOXNGACAMGjRIZt27d++EGTNmCLa2toKamppQtWpVYcKECUJKSopQvXp1QSQSCe/evZPax9raWpBXPAIQrK2ti53O3bt3Cy1atBB0dXWFSpUqCS4uLsLhw4cV7lPQ5ynI+/fvhTVr1giOjo6Crq6uoK6uLlhbWws9e/YUjh49KrVtbGysMGTIEMHa2lpQV1cXbGxshAkTJghv3rwRBg0aJAAQzpw5I7WPs7OzAEB4/Pix1PLCrldWVpYwZ84coWbNmoKamppgamoqfPvtt0JsbKzMPps2bRIACLNmzVL4OXNycoRVq1YJDRs2FLS0tITKlSsLnTt3Fi5cuFDgtbtx44bg6ekpGBgYCCKRSOY8Bd3f06dPC507dxYMDQ0FVVVVwdLSUhg8eLAQFRUls21hebak9zc7O1uYPXu25DoCkLrus2bNEgAImzZtUngMRfc2KSlJmDhxolCjRg1BXV1dsLCwEIYMGSI8e/ZM4XEVHasgJckTgpCXt+fMmSNUq1ZNUFVVFczNzYXRo0cXmF8FQRAiIiKE7t27C/r6+oKmpqbQrFkzYfv27QXeo+zsbOGPP/4Q6tSpI0nb4MGDhZcvXwrt27cXAAj37t2T2S8zM1P43//+Jzg5OQn6+vqCqqqqYGFhITg4OAhz584V7t+/L7PPihUrhAYNGggaGhqS+/nh80VERESK8R2D7xh8x8jDdwy+Y4jxHYOoaESCUIzB6YiIvhJPnz5FtWrVUKtWLaku0kRE8qSkpMDW1haZmZl4+/at3DG8iYiIqGLjOwYRFQffMYjKDucUIaKvWlhYGHJycqSWxcbGYtCgQcjJycGAAQPKKGVEVB7dunVLZuLEpKQkfPvtt3j9+jX69evHlxUiIqIKju8YRFQcfMcgKn/YU4SIvmqNGzfGixcv0LBhQ8kYw9evX0daWhrs7e0RHBwMdXX1sk4mEZUTPj4+OHPmDJo0aQIzMzPExcUhLCwMb968QY0aNXDp0iWOx0tERFTB8R2DiIqD7xhE5Q8nWieir9qYMWOwa9cu3LlzB4mJiVBVVUXt2rXh6+uL8ePH82WFiKQMGDAAWVlZCA8Px6VLl6CkpIRq1aph1KhRmDJlCgwMDMo6iURERFTG+I5BRMXBdwyi8oc9RYiIiIiIiIiIiIiIqELgnCJERERERERERERERFQhMChCREREREREREREREQVAoMiRERERERERERERERUITAoQkREREREREREREREFQKDIkREREREREREREREVCEwKEJERERERERERERERBUCgyJERERERERERERERFQhMChCREREREREREREREQVAoMiRERERERERERERERUITAoQkREREREREREREREFQKDIkREREREREREREREVCEwKEJERERERERERERERBUCgyJERERERERERERERFQhMChCREREREREREREREQVAoMiRERERERERERERERUITAoQkREREREREREREREFQKDIkREREREREREREREVCEwKEJERERERERERERERBUCgyJERERERERERERERFQhMChCREREREREREREREQVAoMiRERERERERERERERUITAoQkREREREREREREREFQKDIkREREREREREREREVCEwKEJERERERERERERERBUCgyJERERERERERERERFQhMChCREREREREREREREQVAoMiRERERERERERERERUITAoQkREREREREREREREFQKDIkREREREREREREREVCEwKEJERERERERERERERBUCgyJERERERERERERERFQhMChCREREREREREREREQVAoMiRERERERERERERERUITAoQkREREREREREREREFQKDIkREREREREREREREVCEwKEJERERERERERERERBUCgyJERERERERERERERFQhMChCREREREREREREREQVAoMiRERERERERERERERUITAoQkREREREREREREREFQKDIkRERERE9FlkpSQhKyWprJNBREREREQVGIMiRERE5YCNjQ1EIlGBf2/fvi3rZBJJ2bx5M0QiEWbPnl3WSflo0dHREIlEcHFxKeukfLWyUpJwfsY4nJ8xrtwHRj51mRwcHAyRSAQ/Pz+p5Z/rmfLz84NIJEJwcLDUchcXF4hEIkRHR3/S8xPNnj0bIpEImzdvLuukEBERUQWkUtYJICIiov94eHjA1NRU7jo1NbXPnJrya/PmzRg8eDBmzZr1VVTIf0rR0dGwtbWFs7OzTAUo0eciDogkPboPADg/YxzazF8BNd1KZZyygrFMpuJgeVs2Zs+ejTlz5mDTpk0ygUYiIiIieRgUISKir8qwcT8gOye7SNuqKKtgw4pfPnGKimf69OlsqU5EX5UPAyIAkPTo/hcRGKloZfLWrVvx7t07WFhYlHVSiIiIiIg+GQ6fRUREX5XsnGxkZ+cU7a+IwRMiIioZeQERMXFgpLwPpVWRWFlZoU6dOlBVVS3rpFA+NjY2sLGxKetkEBEREX01GBQhIiL6guQfhz4uLg4jR46ElZUVVFVVMWHCBMl2UVFRGDp0KKysrKCurg4jIyN4eXnJHc5DPIZ8QX8fjvmdnp6O3377Dc2bN4euri60tbXRrFkzrFy5Ejk5OQrPER0djSNHjqBt27bQ1dWFnp4ePDw8cO3atSJfAxcXFwwePBgAMGfOHKl05h9KKygoCMOGDUO9evWgr68PTU1N1K5dG5MnT0ZCQoLcY4vnEQCAbdu2oXnz5tDS0oKhoSF69OiBqKgoufvl5OTgzz//hJ2dHTQ0NGBubo7hw4cjPj5e4dj9n+M6zp49G7a2tgCAkJAQqWtVnNbvx48fR9u2baGtrQ1DQ0N06tQJ169fL3S/vXv3ws3NDYaGhlBXV0eNGjUwdepUvHnzRuE++/fvR6dOnWBsbAx1dXVYWlqiY8eO2LFjh8y2L1++xHfffYcaNWpAQ0MDBgYGaN++PQICAhQePzIyEr169YKhoSG0tbXRsmVL7N69u9DPcu3aNfTp0wcWFhZQU1ODmZkZ+vfvj3v37slsW9Tn9GtXUEBE7GsIjIhEIoUV1p9jrpqwsDBMmzYNLVq0gImJCdTU1GBpaYl+/frh1q1bxTpWQXOKXLx4Ee7u7tDT04Oenh7atWuH06dPF2l+lBcvXmDw4MEwNTWFhoYG6tati+XLlxeahv3798PBwQE6OjowNjbGwIED8erVKwB55efMmTMlz3+1atWwePFiCIIg97hfSnn7KRU2j0dR5prZs2cP7O3toaOjA319fXh5eSEsLEzhOXfs2IHmzZtDU1MTRkZG6N27Nx4+fKhw+wcPHmDOnDlwdHSEubk51NTUYGJiAm9vb5w/f15mexsbG8yZMwcAMHjw4AJ/u0RGRmLYsGGwsbGBuro6qlSpAm9vb1y+fFlheoiIiOjrxOGziIiIvkDx8fFo0aIF0tPT0bZtWwiCAH19fQDAuXPn0LlzZ6SkpKBu3bpwdHTEs2fPcPjwYRw+fBjLli3DuHHjJMfy9PSUW6EnCAL+/vtvZGZmQllZWbI8ISEB7u7uCAsLg5GRERwdHaGqqopLly5h3LhxOHPmDPz9/SXBhfzWrl2LX3/9Ffb29ujcuTNu3bqFEydO4Pz587h+/Trq1KlT6Gf39PREdnY2QkND0ahRIzRu3FiyLv//+/n5ISMjA/Xr14ebmxsyMzNx8+ZN/PnnnwgICMCVK1dgZGQk9xw//vgjfvvtN7Rt2xZdunTBtWvXEBAQgNDQUNy+fRtVqlSR2n7AgAHYtWsXNDQ04OrqCh0dHQQGBuL06dOoX7++3HN8juvYuHFj9OjRA/v27YOJiQk8PT0lxyjKtQbyKjaHDBkCQRDg4OAAKysrhIWFwdHRURKc+pAgCPDz88PWrVuhqamJFi1awMjICDdv3sTvv/+OwMBAnDt3DsbGxpJ9cnNzMWjQIGzfvh3Kysqwt7eHlZUVYmNjceXKFURERKB///6S7SMiIuDq6oqXL1/C2toaPj4+SEhIQEhICP755x9MnjwZv//+u1S6wsPD4ezsjKSkJNSqVQtNmzZFTEwM+vbti/Hjxyu8BuvWrcPo0aORk5ODZs2awdHREdHR0di5cycCAwNx9OhRtGnTRma/gp7Tr11RAiJiX8pQWuXVggULcODAATRo0AAtW7aEhoYG7t+/j127duHAgQM4duwYnJycPuochw8fho+PD7Kzs9GkSRPUqVMHkZGRcHd3x9ixYwvc9+nTp2jWrBlUVVXh7OyMhIQEnDt3Dt999x2SkpIwc+ZMufutXLkSS5YsQZs2bdCpUydcvXoV27Ztw/Xr13Hp0iV4eHjg3r17cHZ2Rq1atRASEoJp06YhIyMDP//8s9SxvpTytrxbunQpli1bBnt7e3h5eeHOnTsICgrCyZMnceTIEbi6ukptP2fOHMyePRvKyspwdnaGkZERLl68iBYtWqBLly5yz7F27Vr88ccfsLOzQ+PGjaGrq4vHjx8jMDAQhw8fxvbt29GnTx/J9j179sSpU6cQHh4OR0dH1KhRQ7Iu//8HBQXB19cX6enpqFevHry8vPDy5UscPnwYR44cwa5du9CzZ89SvmJERERUXjEoQkRE5cr831cVa3s1VVV8/93wEp0rJydH6nxODi3g5NBSZrtl/9uElLR3MstnTBlTovOWhiNHjqBTp07Ys2cPtLW1JcvT09PRp08fpKSkYN68eZgxY4Zk3dGjR+Ht7Y2JEyeibdu2kgDC9OnT5Z7jxx9/RGZmJlq1aoVevXpJlg8ZMgRhYWEYOnQoli1bJjl/UlISevfujYCAAKxbtw4jRoyQOeaSJUtw8uRJtG/fHkBe5fno0aOxZs0a/Prrr9i0aVOhn3369OkwNTVFaGgofHx8FE60vm7dOnh4eEhdn+zsbMyZMwfz58/HzJkzsWbNGoX7Xrt2DY0aNQIAZGVloWfPnjh06BBWrVqFWbNmSbb9+++/sWvXLpiZmSEkJAQ1a9YEAKSlpcHHxweBgYFyz/E5rqOPjw8aN26Mffv2oU6dOgpbByvy7NkzjBmTl8/9/f3Ro0cPyfmmTp2KP/74Q+5+S5YswdatW9GyZUvs3bsXVlZWAPICHzNnzsTChQvx3XffYdeuXZJ9Fi9ejO3bt8PW1haHDh1CvXr1JOsyMzPxzz//SJ2jf//+ePnyJYYPH45Vq1ZBRSXvZ+3169fRvn17/PHHH3BxcZFUvAmCgIEDByIpKQkTJkzAH3/8ASWlvE7T+/fvV1gZduXKFYwePRr6+vo4ePAgHB0dJeuCgoLQrVs39O/fH1FRUTJDDil6Tr8E/67/E0mPHpRo39zsbCRFP0BORnqR90l6dB/Hhnijkk1NKKmU7BWlUrWaaPjtpBLt+yUbNWoUVqxYATMzM6nlhw4dQo8ePTBixAjcvXtXboV/UaSmpmLw4MHIzs7GypUrJWUCkFeBPXLkyAL337RpE0aOHIkVK1ZIntPQ0FA4OTlh8eLFmDRpktznY/Xq1QgJCZEEHDMyMuDp6YmQkBC0bt0aBgYGePz4MSpVyguk3bx5Ey1atMBvv/2GyZMnSx3zSyhvvwQrVqyQ+i4AgF9//RXTp0/HgAEDEBUVBU1NTQB5vevmzp0LTU1NnDhxQnIfs7KyMHDgQGzbtk3uObp3745Ro0ahevXqUsuvXLkCNzc3jB49Gl27doWWlhYA4Pfff8fs2bMRHh6OYcOGyZ1o/cmTJ+jbty9yc3Nx4MABeHt7S9ZdvnwZnp6eGDJkCFxcXGQaPRAREdHXicNnERFRuXIv8mGx/iKjHpf4XIIgfb74hES52z149ETuuT+Fdu3ayR2+aunSpVLbqamp4X//+59MRdKePXvw4sULNGjQAD/99JPUuo4dO8LPzw85OTlYsWJFgenYunUrfvnlF1hZWeHgwYPQ0NAAkNfSXlxhvWbNGqnzV6pUCZs2bZKkTZ7x48dLKpaAvGFn5s6dCwByh5f6GN27d5e5PioqKpg3bx7Mzc2xb98+hfvOnTtXEhAB8q63uDXzh+lctSovsPbTTz9JAiIAoK2tjRUrVsitiPxSruPGjRvx7t07eHt7S1WCiUQiLFy4UO5kzNnZ2Vi0aBFUVVWlAiIAoKSkhHnz5qFRo0bYu3cvXr9+DSCvkmzx4sUAgJ07d0oFRABAXV0dHTt2lPw7JCRE0uJ7yZIlkopWAGjWrBm+//57AHmVmWJnz57Fv//+C1NTUyxatEgSEAGAbt26oXv37nKvwaJFiyTPTP6ACAB06dIFo0aNwtOnT3H48GGZfRU9p1+CpEcPkHD7Ron+Eu/9W6yAiFhORjoS7/1b4vOWNIhTkKKWyWWpffv2MgERAPDy8kKvXr1w79493L17t8TH37NnD+Lj49G8eXOpgAgAjBgxAq1bty5wfysrK5nn1NHREZ6enkhNTVU4fOKECROkemBpaGhIhp+LiIjAunXrJAERIK+nRqdOnWSO+aWUt1+CHj16SH0XAMD333+PBg0a4MWLF/D395csX7VqFXJzczFixAip+6impoYVK1ZIghofcnBwkAmIAEDLli0xduxYvHnzBmfOnClWupctW4bU1FTMmjVLKiACAK1atcLMmTORkpKC7du3F+u4RERE9OViTxEiIqJyxMPDA6ampjLL7ezspP7dpEkTqcpmsbNnzwIA+vXrJ7cy3s/PD+vXr0dISIjCNJw7dw7ffvstdHV1ERQUBBMTE8m6EydOAMirbFOR05rbzMwMNWvWxK1bt5Ceni5pMSrWqVMnmX2MjIxgaGiIFy9eKExTST158gSHDx9GZGQkUlJSJOPGZ2dnIyEhAW/evIGBgYHMfvLSKR7+JH86379/LxmLvHfv3nL3adKkCW7cuCG1/Eu5juL81LdvX5l1ampq6NWrl0zlcFhYGOLj4yXDX31ISUkJbdq0QXh4OK5duwYPDw9cv34db968Qd26dWFvb1/kdHXv3l1uxZqfnx9++uknXLhwAdnZ2VBRUZHs061bN6irq8vs880330hV6AF5PVtOnjwJFRUVdO3aVW5anJycsGLFCly+fBk+Pj5S6xQ9p/TlKGqZXNaSkpIQFBSE8PBwvHnzBu/fvwcA3L59G0DeXAofBhuLSjyPg7wyDsgrHy5evKhw/3bt2kkC6/nVqVMHR44cUVhm5R9+Skw8HJK1tTXq1q0rs14cmM5/zC+lvAXygrDy5ikSz4MlrxdEmzZtMGzYsFJNhyL5hzAUE4lE6N+/P6ZPn46zZ89iwIABAAr+/jAyMoK7uzsOHDgg9zzp6ek4evQorl27hoSEBGRlZQHIm28EyMvPnTt3LnK6jx8/DgAKg9/i4eU4twgREVHFwaAIERGVK3VqybYOLIjaB8PVFIdIBNSu+d/5jKoYyt2uZjVrucNnfQrTp08v0oSs1tbWcpc/f/4cACSTvX6oWrVqUtt96OHDh+jWrRtycnKwe/duNGjQQGr948d5PXMWLVqERYsWFZjGxMREmZ4EiiqIdXV1kZgov6dOSc2YMUPSyl+R5ORkuUEReenU1dUFkDeUk9jr16+RlZUFTU1NhUNuWFlZyQRFvpTrKM4nivKbvLloxJ/t0qVLhQ7XI67oe/r0KQCgVq1axUqXonxuZmYGDQ0NZGRk4PXr1zAxMSnRZ0lISEBqaioAQEdHp8A0iT9LforO9SWoVK1m4RspUJLhswBAWUPzo4fPKm1FLZPL0v79+zFkyBC8fftW4TbJycklPr644l9RuVNY4K+g8gqQLlPzs7S0lFkmfg7lrcu/Pv8xv5TyFgCOHTtWYKOFLVu2yF3+uYIi8srJ/MufPXsmWVaSMhfIG1rN19e3wIBTcfOzOA8UNreLvHKciIiIvk4MihARUbnyOefpUFZWLtL5vhslfzLpsvRhS9bS8PbtW3Tp0gWvX7/G0qVL5baOzc3NBZA33ERhlQvyWuPnH7LoU/L398eCBQtgbm6OP//8Ew4ODjA2NpakycHBARcvXoQgCHL3/9Tp/FKuo1hx5iIQfzZra+tCK5PLImhQks+irq4uNbGvPK1atZJZ9ime08/lY+fmKM5E6wBQqVrtr3KidXEe+lRiYmLQr18/5OTk4LfffoOXlxcsLS2hpaUFkUiEH3/8Eb/88ovCsu5zKGl5VdB+xTnml1TeKhqOSxxAiI6O/qTn/xT5tThlblpaGrp37464uDj88MMP6Nu3L2xsbKCtrQ0lJSXJvC/Fzc/iz9W/f3+5vYXECssfRERE9PVgUISIiOgrIm7hKm4V+SHx8g9bwmZnZ6Nnz564d+8eRo0ahe+++07u/lWrVgWQNz9J/snGyxvxMEhr166VTLSdX1RUVKmcp3LlylBTU0N6ejoSEhLk9hYR94LI70u5jhYWFrh//z6ePHmCli1byqyXV0En/my2trZFnmhY3BI7MjKyyOkCFOfzly9fIiMjA+rq6qhcubLUPk+ePJG7j7zPUqVKFWhoaCA3Nxfr16+XmUidFFPTrYQ281cUKTDypQdEVFVVkZKSInedvOe/NB0+fBgZGRmYPHkypkyZIrO+NMo6c3NzAIo/y6f+jB/rSylvPwc1NTUAKHF+ffLkidR8W2Li8jP/bwsLCws8evQIT548gbGxscJ98jt37hzi4uLQo0cPLFy4UGZ9SfNz1apVERUVhQULFnzRPfiIiIio9HCidSIioq+IeFzsHTt2yG1JKR56w9nZWWr5mDFjcPr0abi7u2P58uUKj+/u7g4AOHjwYJm2PBZX7GRnZ8tdLx7SRFwZlt/JkycRHx9fKulQVVWV9BDYu3evzPrIyEjcvHlTZvnnvI6FXauCtG3bFgCwe/dumXXv37+XO1l9ixYtYGBggEuXLuHly5dFOk+zZs1gYGCAiIgIXLlypdDtxfk8ICAA797JDm0nzueOjo6SVsHiz7J//365w/Xs3LlTZpmKigpcXV2RlZUldyJ1Kpg4MFKpWm2F23zpAREgL2iQmJgot1wRz2XwqRRU1sXHx+PkyZMffQ7xJNkfzrkj9vfff3/0OT6lL6W8/RzEAa7792UDlfHx8TJDPX5IXjkJALt27QLwX9kMFPz98fr1a7l5s6D8nJmZKfc7Byj8uovzwP79++WuJyIiooqHQREiIvqqqCirQEVFuWh/yl9fh8levXrBzMwMt2/flmllefz4cWzatAnKysoYN26cZPmff/6JdevWwc7ODnv27ClwaIkWLVqgU6dOCAsLw9ChQ+WOpx4ZGamw8qy0iCt2IiIi5K4XD4GxatUqqeFAHj58iJEjR5ZqWsaMyRuCbf78+Xj48KFk+bt37zBu3Di5w5F8zutoZGQEVVVVREVFFbuibsiQIdDU1MSBAwekKpMEQcCMGTMQExMjs4+amhp++OEHZGRkwMfHR+GkwevWrZPaZ+rUqQDyJuX98L5mZmbi2LFjkn87OTmhcePGiI+Px6RJk6Q+V1hYGBYvXgwAmDBhgmS5s7Mz6tevj9jYWPz0009SlaOBgYEKK9tmzpwJZWVljBw5UioNYhkZGdi7d6/UWPr0n4ICI19DQATIm0gcAGbNmiWVr44cOYKlS5d+0nOLy7otW7ZItf5PSUkpdJ6RourVqxeMjIxw+fJlrFmzRmrdhg0bEBoa+tHn+JS+lPL2c3BxcYFIJMK2bdukeuYlJSVh6NChCnuQiPn7+8tMjv77778jPDwcpqam6Nmzp2T56NGjIRKJsGbNGly8eFGy/P379xg/fjzS0tJkji/Oz/7+/oiNjZUsz8rKwrhx4/Do0SO56SrsN8GUKVOgo6ODGTNmYOfOnTLBsezsbBw9ehS3b98u4NMTERHR1+Trqw0iIqIKbcOKX8o6CWVKS0sLf//9Nzp16oQZM2Zg165daNiwIZ49e4bz588DAJYuXYrGjRtL9hFXRpuYmCgcNmvYsGGS1sLbtm1Dx44dsWnTJvj7+6Nx48aoWrUqUlNTcfv2bTx69Aje3t5SlSOlrXXr1jAxMUFAQACcnJxQvXp1KCsro2vXrujatSvGjx+PLVu2YP369QgODkbTpk2RmJiIkJAQ2Nvbw9TUFBcuXCiVtPTu3RsHDx7Erl270KBBA7i6ukJbWxtnz56FhoYGvLy8cOjQIUlLVrHPdR1VVVXRqVMnHDx4EA0bNkSzZs2grq6O2rVrS+69IlZWVli+fDm+/fZbdO/eHY6OjrCyskJYWBgePXqEESNGYO3atTL7TZkyBZGRkdiwYQMaNGiAxo0bo1q1asjNzUVUVBRu3boFHR0dDB8+XLLPtGnT8O+//2L37t1o0KABHBwcYGlpiZcvXyI8PBy6urqS4VZEIhF27twJV1dXrF27FidOnECrVq3w+vVrnDlzBtnZ2Zg8eTK8vLwkxxeJRNi6dStcXFzwxx9/ICgoCE2bNkVMTAxCQ0Mxbtw4ub2k7O3tsW7dOowcORIdO3ZEnTp1ULt2bWhqaiImJgZhYWF49+4dwsLCFE7+XNHJG0rrawmIAHmTse/duxf/+9//EBISAjs7Ozx8+BA3b97EtGnTCp3c+2N4eXmhUaNGCAsLQ7Vq1dC2bVsIgoCzZ89CRUUFgwcPxqZNmz7qHLq6uvjrr7/QrVs3jBo1CuvXr0ft2rURFRWFa9euYezYsVi5cqVMGVeefAnl7edgY2ODoUOHYsOGDWjatKmkZ8eVK1dgZGQEb29vHDx4UOH+Y8aMQbdu3eDg4ABra2vcuXMH//77L9TV1bF161ZoaWlJtm3ZsiV++uknzJ8/H23btoWLiwuMjIxw8eJFJCUlYcCAAdi2bZvU8Zs2bYpOnTrhyJEjqF27NlxcXKChoYHQ0FC8ffsW48aNw4oVK2TS5eHhAU1NTSxduhS3b9+GpaUlRCIRhgwZAgcHB9ja2mLPnj3w9fVF//79MWPGDNjZ2UFPTw8vX77EjRs3kJSUhP3796N+/fqldLWJiIioPGNPESIioq9M27ZtERYWhsGDByMpKQn+/v64c+cOOnXqhNOnT2P8+PFS24t7Mpw5cwZbtmyR+5d/HG9DQ0OcO3cO//vf/9CkSRPcunULe/fuxfXr12Fqaoq5c+dKWup/Kurq6jh69Cg8PT1x+/ZtbNmyBRs3bpQM/VGjRg1cv34dPXv2RFpaGg4ePIinT5/ihx9+wIkTJ0p9boht27bhjz/+gLW1NU6ePImzZ8+iY8eOuHTpkmR4J/HcFmKf8zquX78efn5+SEpKwq5du7Bx48YiDwc1bNgwHDlyBI6Ojrhx4waOHDkCa2trnD9/Hvb29nL3EYlEWL9+PY4cOYLOnTvj2bNn2L9/v2QS4TFjxiAwMFBqHyUlJezcuRO7d+9Gu3btcPv2bezbtw8PHjyAvb09fvlFOuBZt25d3LhxQ5KfAwICcOXKFbRt2xb+/v74/fffZdLVpEkTXL58Gd27d8erV69w4MABpKenY9u2bZg4caLCazBkyBCEhYVh2LBhyMrKwrFjx3DkyBG8fv0aPj4+2Lt3L+zs7Ip0PSuq/D1GvqaACADUrl0bZ8+ehYeHB2JiYnDkyBFoaGjg4MGDGDFixCc9t6qqKs6ePYuJEydCX18fR44cwdWrV+Hj44MbN25I5uv5WF26dEFISAjc3NwQGRmJoKAgaGpq4tixY5IhBD8s48qTL6W8/Rz+97//Yc6cOTAxMcGpU6cQHh6O3r174+LFi9DX1y9w34kTJ2Lnzp14//49Dh48iOjoaHTq1AmhoaFwc3OT2X7evHnYunUrGjVqhPPnz+PkyZNo3rw5rly5gmrVqsk9R0BAAObNmwdLS0ucPHkSISEhaNOmDa5du4amTZvK3cfMzAyHDh1CmzZtcPHiRWzatAkbN26U6g3TsWNH3L59G9999x3U1dXxzz//IDAwEDExMWjXrh02b96MDh06FP1CEhER0RdNJJTlgOBEREREX7GUlBTY2toiMzMTb9++hbKyclkniahMZaUkAcBXExAhYPjw4Vi/fj3+/vtv+Pr6lnVy6BNwcXFBSEgIHj9+DBsbm7JODhEREdFHY08RIiIioo9069Ytmcm7k5KS8O233+L169fo168fAyJEyAuGMCDy5YmNjcWLFy9klu/cuRMbN26EgYEBOnfuXAYpIyIiIiIqPs4pQkRERPSRZs6ciTNnzqBJkyYwMzNDXFwcwsLC8ObNG9SoUUNm0nsioi/J5cuX0aNHDzRq1Ag2NjbIzc1FREQEIiMjoaKigvXr10NbW7usk0lEREREVCQMihARERF9pAEDBiArKwvh4eG4dOkSlJSUUK1aNYwaNQpTpkyBgYFBWSeRiKjEGjVqhCFDhuDs2bM4ffo03r17ByMjI/Tq1QtTp05FixYtyjqJRERERERFxjlFiIiIiIiIiIiIiIioQuCcIkREREREREREREREVCEwKEJERERERERERERERBUCgyJERERERERERERERFQhMChCREREREREREREREQVAoMiRERERERERERERERUITAoQkREREREREREREREFQKDIkREREREREREREREVCEwKEJERERERERERERERBUCgyJERERERERERERERFQhMChCREREREREREREREQVAoMiRERE5cT58+fRp08fWFlZQV1dHZUqVUKNGjXQuXNnLFq0CNHR0WWdxHLBz88PIpEIwcHBZZ0UKSKRCDY2NmWdjBKZPXs2RCIRNm/eXNZJKddcXFwgEon4LFYANjY2EIlEUn9KSkqoXLkyXF1dsWvXLrn7BQcHQyQSwc/P7/MmuIiio6MhEong4uLyyc/1ucpq8b360JdcJtOXhd8NREREXx4GRYiIiMqBRYsWwcnJCX///Tc0NDTg6emJjh07onLlyjh58iR++OEHBAUFlXUyiQrFAAt9TTw8PDBo0CAMGjQIvXv3RvXq1XHmzBn069cPY8aMKevkEX0ymzdvhkgkwuzZs8s6KRUKAyxERESfh0pZJ4CIiOhTysrOQdijOCS9y0IlLTU0rW4CVeXy1SYgLCwMP/74I5SVlbFz50706tVLan1SUhICAgJgampaRiksX3755RdMnz4dVlZWZZ0UIiqGnNxc7LhwEgDwjYMblJTKV1ksz/Tp02V6VQQFBaFr165YvXo1hg0bhiZNmpRN4qhAERERUFVVLetkEBEREVE5xKAIERF9ld7n5GL9iVvYefYeEpLTJcur6Gmin1MdfOveoNwER/bt2wdBENCrVy+ZgAgAVKpUCYMHDy6DlJVPZmZmMDMzK+tkEFExHQ2/hFn7/gIAGGrroUuT1mWcopLp0qUL2rRpg3PnziEkJIRBkXKqTp06ZZ0E+kB0dDRsbW0xaNAg9iYkIiKiMlU+aoOIiIhK0fucXIxd9w+WB4VJBUQAICE5HcuDwjB23T94n5NbRimUFhcXBwAwMTEp1n6FDVOkaDz3/EMz7NmzB/b29tDR0YG+vj68vLwQFham8JyRkZEYNmwYbGxsoK6ujipVqsDb2xuXL1+W2Tb/2PWpqamYNm0aatSoAXV1dfj4+GDFihUQiUQYPny4wvP9+eefEIlEUsPUKPpciYmJmDVrFho0aIBKlSpBW1sbtra28Pb2xt69e2WOnZOTg/Xr16NNmzbQ19eHpqYm6tWrh/nz5yM9PV1mewCIjY3F0KFDYWJiAk1NTTRo0ACrVq2CIAgKP4MiZXH/duzYgebNm0NTUxNGRkbo3bs3Hj58qHD7Bw8eYM6cOXB0dIS5uTnU1NRgYmICb29vnD9/XmZ7GxsbzJkzBwAwePBgqfkYPvycxclLhcnNzcWGDRvg4OAAPT09yb38+eefkZycLLN9/mFhXrx4gcGDB8PU1BQaGhqoW7culi9fXqTzpqeno3LlylBXV0d8fLzcbV6/fg0NDQ3o6uoiJSWl2J/ta5GTm4vlx/dJ/r38uD9yc8tHGVwSxsbGAIDs7Oxi7Xfq1Cl07doVxsbGUFNTg5WVFUaOHInnz5/LbPsx+TQyMhK9evWCoaEhtLW10bJlS+zevbvQ9F27dg19+vSBhYUF1NTUYGZmhv79++PevXtyt8/NzcWKFStQv359aGhowMzMDMOGDZN8rxVXUFAQhg0bhnr16knK5dq1a2Py5MlISEgo1rEKmlNky5YtaNasmaQs7NOnDx49eqSwXM5fHl+8eBGenp7Q19eHlpYWHB0dcfLkyQLTkJOTg8WLF6Nu3brQ1NSEjY0NZs2aJck/T548gZ+fH8zMzKChoYGmTZvi8OHDCj/bixcvMGHCBNSqVQsaGhowMDCAm5sbjh07VmA6cnNzsXTpUtSrVw8aGhowMTGBn58fXr16JbW9i4uLpDHGnDlzpMry8jKcVmHDTBU210x2djYWLlyI2rVrQ0NDA+bm5hgxYoTCvJuZmYm5c+dKfsdYW1tj8uTJSE1NVZjGs2fPYvz48WjcuDGqVKkCdXV12NraYvjw4TLpFv9mCgkJAQDY2tpKXfcPty9OWUJERESyGBQhIqKvzvoTtxBy+1mB24TcfoYNJ259phQVrGrVqgAAf39/vHz58rOdd+nSpejduzdEIhG8vLxgZWWFoKAgtG7dGv/884/M9kFBQWjcuDE2btwIHR0deHl5oU6dOjh8+DDatGkDf39/uedJT0+Hs7Mz1qxZAzs7O3Tt2hWmpqbo06cPVFRUsHfvXmRmZsrdd9u2bQCAgQMHFvhZUlNT0apVK8ydOxeJiYlwcXFBly5dYGFhgX/++Qdr166V2j4jIwMdO3bE8OHDcfv2bTRv3hyenp5ITk7GzJkz0b59e5nAyLNnz9CqVSv89ddfUFNTg7e3N0xMTPDdd99hwoQJBabvUyju/ZszZw6++eYb3Lx5Ew4ODmjfvj0uXbqEFi1a4NGjR3LPsXbtWsyePRtv375F48aN0a1bN1hbWyMwMBAuLi4ylaw9e/ZEo0aNAACOjo6SuRgGDRqEGjVqSLYraV6SJzc3F7169cK3336Lmzdvom3btvDy8kJCQgLmzZsHe3t7hQGLp0+folmzZjh9+jScnZ3h6OiIhw8f4rvvvsO8efMKPbempib8/PyQlZWFTZs2yd1my5YtyMzMRP/+/aGrq1vkz/W1ORp+CQ9e/VcuP3j1DEfCix8AKw+ys7Nx8+ZNAEDdunWLvN/MmTPh5uaGo0ePokaNGvD29kalSpWwdu1aNGvWDPfv35e7X3HzaXh4OFq2bAl/f38YGRmha9euUFNTQ9++fbFkyRKF6Vu3bh3s7e3x999/w8zMDD4+PqhatSp27tyJFi1ayA2EDh06FOPHj0dUVBRcXV3Rtm1bBAYGolWrVnjz5k2Rr42Yn58fdu/eDV1dXbi5ucHNzQ3p6en4888/0aJFC4XPcnH8+OOP8PPzQ3h4OBwdHaXKwsePHxe47+HDh+Hk5IRXr16hY8eOqFevHi5cuICOHTvizJkzCvfr06cP5s6di+rVq8Pd3R1JSUmYO3cuRo0ahaioKLRs2RJnz56Fs7MzWrRogbCwMHh7e8s95pUrV9CwYUMsW7YMubm56NSpExo3bozz58+jY8eOWLp0qcJ0DBgwAD/88AOqVq2Kzp07Q1lZGVu2bIGrq6vU97CnpyccHR0BAI0aNZIqyxs3blzwBf4CiHvnzp07F9WqVYO3tzeAvGegZcuWePHihdT22dnZ6NKlC2bNmoW4uDh07NgRTZs2xYYNG+Dq6oqsrCy555k0aRLWrl0LVVVVtG3bFp07d4aqqirWr18v88zr6Ohg0KBBkgYyPXr0kLruOjo6km1LWpYQERHRfzh8FhERlQsL/S8j4lniRx8nVxDw7+OitSZdfTQc5+89h5KcloRFVdfSED/2bFXi/QHgm2++waJFi/Ds2TPUrFkT3t7eaNOmDZo0aYImTZpATU3to46vyIoVK+Dv748ePXpIlv3666+YPn06BgwYgKioKGhqagLIa8Xat29f5Obm4sCBA5IKBAC4fPkyPD09MWTIELi4uKBKlSpS57ly5QqaNWuGhw8fyqzz9PREUFAQgoKCpNIBALdv38bNmzdRq1YttGpV8DX29/dHVFQUvLy8sH//figrK0vWvXv3DuHh4VLbf//99zh58iS6dOmCzZs3o3LlygDyWoKOHDkSmzdvxty5c/HLL79I9hk9ejRiYmLQs2dPbN++Herq6gCAixcvws3NrcD0fQrFuX/Xrl3D3LlzoampiRMnTqBNmzYAgKysLAwcOFASfPpQ9+7dMWrUKFSvXl1q+ZUrV+Dm5obRo0eja9eu0NLSAgD8/vvvmD17NsLDwzFs2DD4+fnJHPNj8pKi6xAQEABbW1ucOXMG1tbWAIC0tDT06NEDx48fx6hRo+QGWjZt2oSRI0dixYoVUFHJ+1kcGhoKJycnLF68GJMmTYK2tnaB5x81ahSWLFmCdevWYerUqTItk9etWwcAGDlyZKGfpTyZu38z7j5/UirHEgQBt2JkA2/f7/oftp0/Lrc1d0nYWVjj525+pXIsebKysvDw4UMsWLAADx8+RJMmTeDp6VmkfQMCAjB//nzUqFED+/fvR/369SXr1q5di5EjR8LPzw8XL16U2bc4+VQQBAwcOBBJSUmYMGEC/vjjD8ncLfv370fPnj3lpu/KlSsYPXo09PX1cfDgQUmFOJAXxOzWrRv69++PqKgoyTwdAQEB2Lx5M4yNjRESEiIZriolJQVeXl4IDAws0rXJb926dfDw8JB67rKzszFnzhzMnz8fM2fOxJo1a4p9XLFLly5h0aJF0NbWxunTpyXfLe/fv8fQoUOxdevWAvf/448/sGXLFgwYMECybNGiRfjhhx8wd+5ctGvXTmafJ0+eQENDA5GRkTA3NwcAxMTEoEmTJvjrr78QGhqKvn374o8//pB8d61atQpjx46VOWZKSgq6deuGxMRErFmzBsOHD5c8P/fv34eHhwemTJkCNzc31KtXTyYdKioquHfvnqScTE5ORocOHXD16lXs3r0bgwYNApA3l46pqSlCQ0Ph4+NTbnqHlJanT58iMzMTN2/elOTbjIwM9O3bFwcOHMDYsWMREBAg2X758uU4deoUateujTNnzkiG8Hz58iVcXV0REREh9zxz5syBg4MDDAwMJMsEQcC6deswcuRIfPfdd5LePVWqVMHmzZvh4uKCV69e4ffff5fb0+ljyhIiIiL6D3uKEBFRuRDxLBFXH7z66L/rUXFFHhbrfU4urkfFfdT5SiOQY2tri8OHD8PGxgapqanYsWMHRo0aBXt7e+jr68PX1xe3bpV+r5YePXrIBCK+//57NGjQAC9evJCqRF62bBlSU1Mxa9YsqUpsAGjVqhVmzpyJlJQUbN++Xe65Vq5cKbeCW9wDRF6lvHhZ/sonRcSth11dXaUCIgCgpaWF1q1bS227du1aGBkZYfv27ZKACACoq6tj1apVMDExwbp16yTD+0RHRyMoKAgaGhpYuXKlJCACAK1bt8aoUaMKTWNpK879W7VqFXJzczFixAhJQAQA1NTUsGLFCklQ40MODg4yAREAaNmyJcaOHYs3b94U2Dpano/NSx8St4pevHixpKIPALS1tSUtdAMCAvDkiWwFv5WVFZYsWSKpaAbyerh4enoiNTUV165dK/T8NWrUgJubGx4+fIhTp05JrQsODsb9+/fRqlWrL6519d3nT3D54d1S+bvyKALp72V7g6W/z8SVRxGldp7SCuLk165dO8nwNerq6rCzs8POnTsxZswYnDlzRqa8UWT+/PkAgK1bt0pVYgLAiBEj4OXlhUuXLkl6oORXnHx69uxZ/PvvvzA1NcWiRYukJrPv1q0bunfvLjd9ixYtQk5ODlasWCEVEAHy5lAZNWoUnj59KjWk04oVKwDk9bzIP3+Hrq4uVq5cWaJgV/fu3WUCkSoqKpg3bx7Mzc2xb98+BXsWzerVqyEIAkaNGiUVbFdVVcXSpUulWuPL07NnT5nvpEmTJkFfXx8XLlzA+/fv5e63fPlySUAEyOsh+s033yA3Nxfp6elYvHixVF4aMWIEDA0NERoaKnXMTZs24cWLF/j2228xYsQIqWtcu3Zt/Pnnn5KhIRWlI385qaenh6lTpwKAzFCNX7uff/5ZKt9qaGhg1apVUFdXx4EDB6S+M8R5/ddff5Wa08zU1BS//fabwnN07txZKiAC5A3fNWLECDg4OODkyZPFHlbxY8oSIiIi+g+DIkREROWAi4sLIiMjERgYiLFjx6JFixZQU1NDeno69u7di+bNm+PAgQOles7+/fvLLBOJRJLlZ8+elSw/fvw4ACisUHNycgIAufNBmJiYwN7eXu5+Xl5eqFSpEo4cOYLXr19Llufm5mLHjh0QiURFCoo0a9YMQF7F+K5du+TOIyEWHByMrKwstG/fHpUqVZJZr6WlhebNmyMxMREPHjwAAJw7dw6CIKBdu3Zy53755ptvCk1jaSvO/RP/f9++fWX2MTIygru7u8LzpKenIyAgAD/++COGDx8OPz8/+Pn5SSrQIiMji5Xuj8lLH4qJiUF0dDQ0NTXlHs/a2houLi4QBAHnzp2TWd+uXTtoaGjILBdXlH04hIoio0ePBgCZYdrE//7SeonQfzw8PCTD1wwcOBBubm6oVKkS1q1bh59++knhsDn5xcXFISwsDGZmZlIB2vwKyvfFyafiZ71bt25SwVsxeWVVbm4uTp48CRUVFXTt2rVI6cvOzpa0RJdXrtSvX18ylF5xPXnyBKtXr8aECRMwdOhQSZmTnZ2NhISEEg3LJSYeAqx3794y6wwNDQssCwGgU6dOMsvU1NRQrVo1ZGVlyZ33RFVVFe3bt5dZLh5SsF27djI9QlVUVGBra4v3799LHfNjyk8VFRW5n6+45V1Rie9b/r8pU6YAyLsP8tbLG6LtU5H3HWpubo527dpBEARJWsTfM9ra2nKfD3mBj/zi4uKwceNGTJ48WdKD0s/PDy9fvkRubi6ioqKKnOaPLUuIiIjoPxw+i4iIyoW6loalcpzkd1m4/7zoFSa1LQygp1Xy4alKK91AXsWJl5cXvLy8AOTNk3Hw4EFMnz4dz549w9ChQ+Hm5lbocD5FpWgCWvHyZ8/+G/9fPM56/laV8sirEMrfKvVDGhoa6NWrFzZs2IC///5bUrl85swZPH/+HM7OzgXuL+bq6oqpU6fijz/+QL9+/aCsrIx69eqhXbt2+Oabb9C8eXOZz7J79+5CJx5OSEhA7dq1JROXKkqLomv5KRXn/pU0/aGhofD19S2wsqygAJQ8H5OXPpT/c+VvEZ9ftWrVpLbNz8rKSu4+4rk/FM1186EuXbrAysoKBw8exMuXL2Fqaor4+HgEBATAwMBAbgVseWdnUfhzVxSvU5MQ9argiX9rmFigso5sgLK4SivN+U2fPh0uLi5Sy5KTk9G7d29JD6zVq1cXeAzxBMmxsbGF9p6Ql++Lk09L8qwnJCRIJosurKeEOH0JCQnIzMyEpqamZNJ5eecqbmv1GTNmSHqtKJKcnFxgJXRBxGWZomuqaHlh6wsqM0xNTeX2KBJfa0tLS7nHFK/Pf0xx+VnYsG3y8pGZmZlUb6OipP1jbNmyReG6hw8f4uHDhzLLXVxcpHozfir6+vpyG0UAst+h4meqatWqCp9fa2trucG61atXY/LkycjIyFCYluJ8h35sWUJERET/YVCEiIjKhY+dl0MsKzsHrjP9kZCcXui2VfQ04T/NC6rK5bPjpI6ODvr3748GDRqgUaNGSExMRGhoaKEtWcXEQz+VBvGx+vfvL7dSRUxeRbd4XgtFBg4ciA0bNmDbtm2SoEhxhs4SW7x4MUaOHInAwED8888/OH/+PJYtW4Zly5bhhx9+wMKFC6U+S7169aSCJfLkH1oLQKnNfVAUpXn/xIqT/rS0NHTv3h1xcXH44Ycf0LdvX9jY2EBbWxtKSkpYt24dRowYAUEQipWGj8lLpU1RIKW4lJWVMXz4cMyYMQN//fUXfvzxR2zevBlZWVkYNWpUoc9AeVQac3Pk5Oai4+KphW4nggg7R88stfvxqenp6eG3337DsWPHsG7dOixcuBD6+voKtxfn+cqVK6NLly4FHvvDeSCAkuXT4jzr4vSpq6ujT58+BW5b2PxOH8Pf3x8LFiyAubk5/vzzTzg4OMDY2FjS48XBwQEXL14sdplTmkpyLwrbpzjHFN8rb2/vAvOcvOEqP/fzJe8+RUdHw9bWFoMGDcLmzZs/6flL+zu0uN//V69exdixY6Gjo4MVK1bA1dUVZmZmku+Dfv36YdeuXcXKzx9blhAREdF/GBQhIqKvipqKMvo51cHyoLBCt+3vVKfcBkTya9iwISpXrozXr19LtfwTD7ehaDzqp0+fFnjcJ0+eyB3eRNwS0cLCQrKsatWqiIqKwoIFC4rUc6M42rRpAxsbG1y6dAlRUVGwsLBAQEAANDU10atXr2Idq1q1apgwYQImTJiAnJwcBAQEYODAgVi0aBEGDBiAunXromrVqgCAFi1aYNOmTUU6rvhayJuXAvjvmhXH57x/FhYWePToEZ48eSK3Vbe89J87dw5xcXHo0aOHJKCUX3GG/MivNPNS/vuSm5srt9JP3LI6//X4FIYNG4a5c+di/fr1mDZtmmRM/xEjRnzS85ZnR8Mv4cGrZ4Vu9+DVMxwJv4wuTeQPB1MeiXsg5eTkICoqqsAAq7jM0dXV/eQVwSUpq6pUqQINDQ3k5uZi/fr1konUC1K5cmWoq6sjPT0d8fHxMDIyKtK5CiKeB2nt2rVyK3xLWubkZ25ujsePH+Pp06dyy8LCyt2yVrVqVdy/fx8TJ06Es7NzWSenTBX0HZqdnY3Y2FiF+759+xbJycnQ09OTWffhd6j4v0+fPoUgCHKDI/Ket3379kEQBCxcuBDDhg2TWV+S/Pw5yxIiIqKvXfmvCSIiIiqmb90bwLm+/OEoxJzrW2KYe4PPlKKCFdZKMDExEYmJeRO65x9mQzxp6/3792X2iY+Px40bNwo87s6dO+Uu37VrF4D/xqUGIOmdsn///gKPWRIikUgyzv327dtx4MABpKSkwNvbW26FRVEpKyujV69ekvHBb9++DSBvqC0VFRUcO3YM6emF9ygCIBnO459//kFcXJzMekXXsiCf8/61bdsWAOQOF/b69WucPHlSZrk4z4krYfLLzMxUOOGxuKIqOztb7vrSzEtVq1aFjY2NZN6TDz19+hRnzpyBSCSSXINPxcTEBN27d0d0dDSmTZuGBw8ewMnJCXXr1v2k5y2vcnJzsfx40SfFXn7c/5P0jvpU8g/9U9iQhhYWFrCzs0N0dPQnn/xYnM/3798vdzgkeeWGiooKXF1dkZWVJTWRekFUVVUlc0XJK1fu3r2L8PDw4iS9wDLn5MmTiI+PL9bx5BGX5Xv37pVZ9+bNG5w4ceKjz/Epfcrv4g8VVpaXtYK+Q0+fPl1ouuU9Cy9fvpR8Z4jzStWqVWFtbY20tDQEBQXJ7HP06FG5Q2cVlJ8jIiIQFia/8U5B1/1zliVERERfOwZFiIjoq6OqrISVw13xXZcmqKInPWxNFT1NfNelCVYOdy03vURmzJiByZMny52w+vXr1xgwYAAEQUDVqlWlJtZ0cXGBSCTCtm3bpPZNSkrC0KFDFfZAEPP395eZvP33339HeHg4TE1N0bNnT8nyKVOmQEdHBzNmzMDOnTtlAjnZ2dk4evSoJPBQXOJhsrZv346tW7dKLSuK/fv3IzQ0VCZdsbGxkooH8VjwZmZmGD58OF6+fIlevXpJzb0h9uzZM0k6AMDW1hadOnVCRkYGxo8fLzW58uXLlwudU0Cez3n/Ro8eDZFIhDVr1kgmRwaA9+/fY/z48UhLS5M5vnj4Kn9/f6kWt1lZWRg3bhwePXokN13iiqqIiAi560s7L3333XcAgGnTpkm18n737h1GjRqF9+/fo3v37qXew0ke8fBvf/zxB4CKPcH6iVtX8SThJdSUVYr09yThJY7fulrWyS6S5ORkTJ2aNyxYjRo1ijTU2+zZswHkTfAtbwLklJQUbNy4sciBWkWcnZ1Rv359xMbG4qeffpJ6vgIDAxUGM2fOnAllZWWMHDkSx44dk1mfkZGBvXv3SpWXY8aMAQAsWLBAqgxLTU3F2LFjiz3Mlfg6iudqEXv48GGpPUvisnD16tW4evW//JadnY1JkyZJ5lYpr0aMGAEzMzOsXLkSy5Ytk6k4F08QHhoa+tHnKqwsL2vt2rUDkDd0Zv7vy/v372PcuHGF7j937lypfJuRkYGxY8ciMzMTXl5eUvPviPP6tGnT8PLlS8nyuLg4SVnwIXF+XrdundRvhri4OAwaNEhh0Kaw6/65yhIiIqKvnkBERPQVy8rOES7dfyEcD4sWLt1/IWRl55R1kmR89913AgABgFC9enXB29tb6Nu3r+Ds7CxoamoKAAQ9PT3h3LlzMvsOGzZMACBoa2sLHTt2FDp27ChUrlxZqFOnjuDt7S0AEM6cOSO1j7OzswBAGD9+vABAcHBwEPr27Ss0bNhQACCoq6sLJ06ckDnXkSNHBB0dHQGAYGtrK3Tu3Fno27ev0K5dO6FSpUoCAGH//v2S7R8/fiwAEJydnYt0HVq1aiW5DiYmJkJ2drbc7QYNGiTzucTX0NTUVOjYsaPQv39/wcPDQ9DS0hIACD179pQ6RkZGhtC5c2cBgKChoSG0bt1a6NOnj9CtWzehbt26AgChUaNGUvs8efJEsLCwEAAIVatWFfr06SN06NBBUFFRkVxLa2vrIn1Wsc95/2bMmCEAEJSVlYX27dsLffr0EaytrQV9fX1hwIABAgBh06ZNUvt06tRJACDo6uoKXl5eQq9evQRzc3NBS0tLGDdunABAmDVrltQ+L168EDQ1NQUlJSXB3d1dGDJkiDB06FAhNDRUsk1x81JBcnJyhG7dugkABC0tLaFLly5Cr169BBMTEwGAULduXeHVq1dS+2zatElu2sVmzZol93qIr/3jx48VpqdBgwYCAMHIyEjIzMws0meg8sfa2loAIHh4eAiDBg0SBg0aJAwcOFBwd3cXDAwMJM9F/nwtCIJw5swZAYAwaNAgmWPOnTtXEIlEkvKlR48egq+vr9CyZUtBTU1NACC8efNGsn1J8+mNGzcEPT09AYBQu3ZtoW/fvkKbNm0EkUgkKTfklcsbN24UVFVVBQCSMqhPnz6Co6OjpCwNCwuT2kdcdmhoaAidOnUSevXqJRgbGws2NjaCl5eX3DJMkQcPHkjSXbNmTaF3796Cm5uboKamJjg5OQkODg5ynz/xvfqQojJ5+vTpAgBBRUVF6NChg9CnTx/BxsZGMDAwkHyeHTt2SO0j73snP0VlQ0HfC4XdX0XHvHr1qqR8Mzc3F9zd3YV+/foJ7u7ugpGRkQBAWLJkSZHToei7OiMjQ3Ketm3bCn5+fsLQoUOFgwcPyj1OUYjPJe/5KK6MjAyhfv36AgDBzMxM6N69u+Dk5CRoaGgIffv2LTBfWFlZCd7e3oK6urrQsWNHwdfXVzA3N5esi4mJkdrn/fv3Qrt27STPvY+Pj9CtWzdBT09PaN68udC6dWuZe5WYmChYWloKAARLS0uhZ8+eQufOnQUdHR2hdu3ago+Pj9w8tX//fsl5evToIQwdOlQYOnSokJCQINmmuGUJERERyWJQhIiIqIzFx8cLO3bsEPz8/IRGjRoJRkZGgrKysqCnpyc0a9ZM+OGHH4Tnz5/L3ff9+/fCnDlzhGrVqgmqqqqCubm5MHr0aOHNmzcKK3HyV7Ts3LlTaNGihaClpSXo6ekJnTp1Eq5du6YwrdHR0cJ3330n1KlTR9DU1BS0tbWFGjVqCD4+PsLmzZuFlJQUybbFDYqsXLlSEhSZOHGiwu3kfa6wsDBh6tSpgr29vWBqaiqoqakJ5ubmgouLi7Bjxw65AZbc3Fxh586dgru7u1ClShVBVVVVMDU1FZo3by5Mnz5duHHjhsw+z549E/z8/AQjIyNBXV1dsLOzE5YtWybk5uaWKCjyue/f1q1bhaZNmwrq6upC5cqVhR49egiRkZEKK1czMjKEefPmCXXr1hU0NDQEY2NjwdfXV7h7926BFXqnTp0SnJycBF1dXUnFzYfHLk5eKkxOTo6wbt06wd7eXtDR0RHU1dWFunXrCjNmzBCSkpJktv+UQZEJEyYIAITvv/++yOmn8kdcofrhn5aWlmBnZyd89913wtOnT2X2KygoIgiCcOHCBaFv376CpaWloKamJhgYGAj169cXhg4dKhw+fFjIzc2VbFvSfCoIghARESF0795d0NfXFzQ1NYVmzZoJ27dvL7Rcvn37tjBs2DChWrVqgrq6uqCnpyfUqVNH6Nevn7B3716ZQF92drawdOlSoW7duoKamppgYmIi+Pn5CbGxsYUGEuR58OCB0LNnT8Hc3FzQ0NAQateuLcyaNUvIyMhQ+PwVNygiCHnXtnHjxpKysGfPnsKDBw+EoUOHCgCEY8eOSW1fnoIigiAIcXFxwowZM4RGjRoJ2tragqampmBrayt4enoKq1evFuLj44ucjoLyxI0bNwRPT0/BwMBAUpYrSm9RlGZQRBAE4dWrV8LgwYMFY2NjQU1NTahdu7bw22+/CTk5OYXmi6ysLGHOnDlCzZo1BTU1NcHU1FT49ttvhdjYWLnnSk9PF37++WfB1tZWUFNTEywtLYXx48cLycnJCu9VbGysMGTIEMHa2lpQV1cXbGxshAkTJhT4HS8IgrBixQqhQYMGgoaGhqTs+fDYxSlLiIiISJZIEIrZr5mIiIi+aC4uLggJCcHjx4+lhoegLwPvX/mVk5MDGxsbPH/+HA8ePED16tXLOklEVEQ5OTlo0KAB7t27h9jYWJiYmJR1kugTEIlEsLa2lkyoTkRERBVT+RhMnYiIiIjoC7dp0yY8e/YMXbp0YUCEqJy6f/++zNwhGRkZmDx5MiIiIuDm5saACBEREdFXTqWsE0BERERE9KV6/fo1pk2bhvj4eBw5cgQqKiqYN29eWSeLiBT43//+h3Xr1qFp06awtLTEmzdvEB4ejlevXsHY2BgrV64s6yQSERER0SfGoAgRERERUQmlpKRg48aNUFVVhZ2dHebPn49GjRqVdbKISAFvb2/ExMTg6tWruHHjBgRBgKWlJXx9fTFt2jRYWFiUdRKJiIiI6BPjnCJERERERERERERERFQhcE4RIiIiIiIiIiIiIiKqEBgUISIiIiIiIiIiIiKiCoFBESIiIiIiIiIiIiIiqhAYFCEiIiIiIiIiIiIiogqBQREiIiIiIiIiIiIiIqoQGBQhIiIiIiIiIiIiIqIKgUERIiIiIiIiIiIiIiKqEBgUISIiIiIiIiIiIiKiCoFBESIiIiIiIiIiIiIiqhAYFCEiIiIiIiIiIiIiogqBQREiIiIiIiIiIiIiIqoQGBQhIiIiIiIiIiIiIqIKgUERIiIiIiIiIiIiIiKqEBgUISIiIiIiIiIiIiKiCoFBESIiIiIiIiIiIiIiqhAYFCEiIiIiIiIiIiIiogqBQREiIiIiIiIiIiIiIqoQGBQhIiLKZ/bs2RgzZkxZJ4M+UnBwMHx9fXHnzp1i7ztmzBjMnj279BNFRJ9MXFwcfH19sWfPnrJOSpnx9fXFqlWrSrTvx5SZ8vC7lL42pf2MFAWfIyIiok9HpawTQEREX6c7d+5gzpw5kn8rKSlBS0sLRkZGqFmzJlxcXFCjRo0yTCEVZM+ePbCysoK9vX1ZJ0WhuLg4BAcHo1mzZqhevXpZJ+eT+xT3JCMjA4GBgXj06BEePnyIpKQkODs7l1olzK1btzBv3jwAwJIlS2BhYVEqx/3wHIcOHUJMTAySkpKgo6MDExMT1K1bF927d4eGhkapn5OKrzw9r1euXEF0dDR69OgBZWXlT3qu4OBgJCcno2vXrkXavjxdJ/o4giDg3LlzuHr1Kh4/fow3b97A0NAQtWvXRq9evWBiYlLiY1+6dAlHjx5FdHQ0cnNzYWxsDGdn5yLns6KIjo7GlStX4OTkBFNT01I7LpVPxS2riIiIvnQMihAR0Sfl4uKC+vXrQxAEvHv3Dk+ePEFoaChOnDgBNzc3DB06FEpK7LhY3vj7+8PBwaFcB0Xi4+Ph7+8PfX19mcpDJycnODg4QEWl+D91li5dCpFIVFrJLDWf4p4kJyfD398fBgYGqF69Om7cuFFqx37//j02bNgAdXV1ZGZmltpx8zt69Cg2bdqEqlWrws3NDfr6+khMTER0dDSOHj2KDh06MChSThT0vH5uV69eRUhICHx8fD5LUOT58+dFrmj8mOu0ffv2cvN9OmPGDAiCUNbJKFPv37/HypUrUa1aNTg5OaFKlSp49eoVTpw4gStXrmDu3LmwsbEp9nG3b9+OQ4cOwd7eHv3794dIJMKrV68QHx9fqumPjo6Gv78/atWqxaBIBVDcsoqIiOhLx6AIERF9UjVq1ICTk5PUsoEDB2LJkiU4efIkDAwM0LNnzzJKHX2J0tPToampWeA2SkpKUFNTK9HxVVVVS7Tfl8jAwABr1qyBoaEhsrKy8M0335TasQ8ePIh3796hffv2OHLkSKkdVywnJwd79uyBtbU1Fi5cKHPf0tLSKtS9pIopOzsbubm5UFNTK3GZ9ymUJCBdEoIgIDMzs1wGP5WVlTF79mzY2dlJLXdwcMD06dPh7++PKVOmFOuYYWFhCAwMxPjx49GmTZvSTO5Xqyi/GYiIiKjiYVCEiIg+Oy0tLUycOBFjxozBoUOH0LlzZ6kX1vv37yMgIACRkZHIysqCubk5OnbsCFdXVwB5laEjR45E1apV8fPPP8scf+XKlbh48SLWrl0LHR0dAEBCQgL27t2LmzdvIjk5GYaGhnB0dETPnj2LVJEUGhqKoKAgxMTEQFlZGbVr14avr6/UEGBxcXEYO3YsevbsCSMjIxw8eBDx8fEwNjZGt27dZIJDvr6+cHZ2hqOjI3bv3o2YmBhUqVIFffr0QevWrfHo0SNs27YNUVFR0NLSQteuXdG5c2eZtF2/fh2HDh3Co0ePkJubC2tra3Tr1g3NmzeXmzZra2v4+/vj+fPnMDAwgLe3N9zd3aW2A4ALFy7gwoULAAAjI6MCx6rfs2cP/P398dtvv+HYsWO4cuUKsrKyYGdnBz8/P5iZmUm2TUxMRGBgIG7fvo34+Hjk5OSgatWq6Ny5s0wlz6pVqxASEoJ169Zh69atCA8PR0pKCkaPHo3Vq1cDADZs2IANGzYAgGTop+DgYKxevRqzZs1CvXr1JMdLSUlBQEAArl27htevX0NbWxvVq1eHr68vqlWrBiBvThEjIyOpeUXE98re3h5///03nj9/Dn19fXh6esLLy0smr5w7dw7R0dFISkqCnp4eGjVqhH79+kFfX/+T3JN//vkHx48fx8uXLyEIAgwMDFC3bl2MHDlS4T0D8gJAhoaGBW5TEi9fvsT+/fsxYsQIvHz5stSPD+Tdy7S0NLRp00Zu8ENbW1vq38+fP8fhw4dx9+5dvH79GiKRCLa2tujRowcaNmwote3s2bMRHx+PmTNnYtOmTYiIiIC6ujo6dOgAX19fpKSkYPPmzQgLC0Nubi5at26NoUOHyqQjJiYGe/fuxZ07d5Ceng4jIyO4urrCy8ur0Bb96enpOHDgAMLDw/Hq1StkZWXB1NQUbm5u8PDwkOnNVJS8DeSVr/v370dkZCQyMzNRuXJlNGrUCIMGDZJUZOfm5uLIkSMIDg5GbGwsVFVVYWdnh759+6Jq1aqSY4mfs59++gm3bt3CuXPnkJqaiurVq2PgwIGS8lG8HSD/eQWAzMxMHDx4EKGhoYiPj4empqbkualSpYrUZ33y5Am2bNmCyMhIaGhowN7eXvK8FGb27Nm4e/cuAEgFAfOXFUW9b3fv3sW+ffsQHR2NjIwM6OnpoUaNGhgwYACMjY0xZswYSet9X19fyX4rV66EsbGxTNoKu075r3dERATOnj2L169f4+eff0a9evUk5ZT4mubm5mL//v24efMmXrx4Ifksbdq0Qbdu3QoNXCQnJ2Pv3r24ceMG3rx5A01NTZibm8PT0xOOjo6FXuf4+Hip7w1x2TpkyBDJ/VNTU4OjoyMGDhxYaHryl5mmpqYIDAzEixcv4OPjA19fX7llN/Df91P+6y7+bvnrr7+wdetWXL16FdnZ2ahXrx6GDRsmledKeh2UlZVlAiIAYG1tDTMzM8TExBT4eeU5cOAAqlWrJvmuTE9Ph4aGRpF7NxY1T4ivGQAsXLhQsn/Pnj2l8rI8ZfkbTpzvZsyYga1btyIiIgLa2toYOnQoFi1ahLFjx8r8FktLS8Pw4cNhb2+PcePGFXoNs7OzsX37doXlHQCFv0EAyDynQF75t2vXLoSGhiI9PR3Vq1fHgAEDFKbhyJEjOHr0KBITE2FqagofHx/ExsbK5POiXrvillVERERfAwZFiIioTGhpaaFFixYIDg7G/fv30bhxYwB5Y70vWbIEtra26NatG9TU1HD9+nWsWbMGb9++Rffu3aGsrIzWrVvj+PHjSExMlKrUzczMxJUrV9CkSRPJy/SrV68wY8YMKCsro3379jA0NERUVBQCAwPx+PFj/PjjjwVWKAQFBWHr1q2wtbVFnz59kJGRgZMnT2LWrFmYOXMm6tSpI7X91atXkZiYCA8PD2hpaSEkJAQrV65Ebm4uXFxcpLZ9/Pgxbt68CTc3Nzg5OeH48eNYunQpAOCvv/6SVMQHBwdjy5YtsLa2Rv369SX7i4cPatiwIXr37g2RSITQ0FD89ttvGDt2LNq2bSt1vrCwMJw+fRpubm7Q1tbGP//8gw0bNsDCwgL16tWDnp4exo4di5UrV6JmzZrw8PAAgCK3wl25ciW0tLTQo0cPJCYm4tixY5g1axZ+++03VKpUCQDw9OlTXL9+HS1btoSpqSnS09Nx7tw5LF++HNnZ2TLXCAAWLFiAKlWqSCqk69ati27dumH//v2SIdoAFDjER3JyMn766SfExcXB2dkZ1atXR3p6OiIiInD//n2pimN5Hj16hCtXrsDNzQ0uLi64fPkytm3bhszMTKneTidOnICenh48PT2ho6ODx48f48yZM3jw4AEWL14sU3H+sfckODgYa9asQbNmzeDq6gqRSIS4uDhcvXq18Bv2iWzcuBHVq1eHk5PTJ5v4Wk9PD+rq6rh58yZev36NypUrF7j9nTt38ODBA7Rq1QpGRkZITk7G6dOnsWDBAsycOVPquQLy5luZN28emjRpgmbNmuHKlSvYt28fNDU1cf78eVhaWqJv3764e/cu/vnnHxgYGKB3796S/SMjIzFv3jxUrlwZXbp0gba2Nu7cuYMdO3bg5cuXGDFiRIHpTUxMRHBwMFq1agUXFxfk5OTgypUr+Ouvv5CSkoJevXpJti1q3r506RKWLVsGPT09uLu7o0qVKoiLi8Ply5eRmZkpqRBdunQprl69irZt28Ld3R2pqak4fvw4ZsyYgYULF8rMDbN9+3YAgJeXF969e4fjx49j7ty5WLhwISwtLQt9XrOzszF//nxER0ejXbt2sLKyQkJCAo4fP467d+/i119/lZQfL1++xKxZswAAXbp0gZ6eHi5cuFDkCca7d+8OQRAQERGBUaNGSYbPEn+mot63Z8+eYcGCBTA3N4ePjw+0tbWRmJiIf//9FwkJCTA2NsagQYOwa9cuvHnzBkOGDJGkQU9PT27ailquia+3p6cnVFRUYGBgIPd42dnZCAoKQuvWrdG6dWsoKSkhPDwce/fuRXx8PEaPHl3gtfrzzz/x6NEjuLu7w9zcHGlpaXjy5Anu379faFBEkcTERMyfPx8ODg5o1aoVbt68iWPHjkFHR6fQynaxK1eu4M2bN/Dw8IChoWGhz35BFixYACMjI/Tt2xcvXrzAsWPHsHLlSqnASmlfh5ycHCQlJRV7jqWMjAzcv38fbm5u2LNnD44ePYq0tDRoa2vD2dkZ/fv3L7R3XFHzRKtWrfD27VucOnUKXbt2hZWVFYC8gE5BysNvuIyMDMyZMwf169dH//79kZmZicaNG8PQ0BAhISEyQZHQ0FC8f/8e7dq1K9J9KKy8K4klS5bgxo0baN26Nezs7PDkyRMsWLBAcg3yCwgIwO7du1GrVi107NgRqamp2Lx5s0zwGCj6tStuWUVERPQ1YFCEiIjKjPgl+8WLF2jcuDGysrKwdu1aNGrUCNOmTZO85Hp6euKPP/7Avn374ObmBl1dXTg5OeHYsWMIDQ2Vaql/9epVZGRkSAUD/vrrL6ioqGDx4sXQ1dUFAHTo0AE2Njb466+/EBYWhqZNm8pNY2pqKnbv3g1ra2vMmzdP0qrOxcUFEydOxF9//YXFixdL7RMTE4PFixdLPl+HDh0wdepUbN++HW3atJFqDSveVlzR0KBBA0yePBlLly7F9OnT0aRJEwB5w22MHDkSJ0+elFSUJSYmYtu2bXB3d8ewYcMkx/T09MSMGTOwbds2ODo6SrVsfvbsGZYtWyaphGjdujVGjx6NEydOoF69etDQ0ICTkxNWrlwJIyMjmcqDwmhoaODnn3+WVDTWq1cPv/zyCwICAjB48GAAgJ2dHZYvXy5VieHp6Ylp06ZJKgM/VK1aNYwaNUpqWcOGDbF//365Q7TJs3PnTrx69QqTJ09Gq1atJMt9fHyKNPZ9TEwMfvjhB8k98fDwwOzZsxEQEIAOHTpIeoH8+OOPUFdXl9q3Zs2aWL16NS5fvizTG+Zj78nVq1dhYWGBadOmSS0vzaGwiuPChQu4desWfv311xLtn5ycjJSUFFSpUkXmOuanpKSEnj17YseOHRg7dixq1qyJWrVqoVatWmjQoIHMcCnOzs4yvQk6dOiASZMm4eDBgzJBEXHgwdPTEwDg6uqKMWPGYMeOHfDy8pJcX3d3d7x69QqnTp2SBEUEQcDatWthbm6O+fPnSyoq3d3dUblyZQQFBaFTp05SvS4+ZGJigtWrV0vNedGpUyfMmzcPQUFBUq26i5K3MzIysHbtWhgaGmLRokWSshAA+vXrJ9nu4sWLuHTpEiZMmAAHBwfJNuIy7++//8akSZOk0pqRkYHffvtNEqizt7fH1KlTsXPnTnz//fcwMTEp8Hk9cuQIHjx4gLlz56JWrVqS5fb29pg+fToCAwMlraZ3796N9PR0/PLLL5Jgj7u7uyRQUpiGDRvi3LlziIiIgKOjo1Qr8+Lct3///Rfv37/HjBkzJAEbAFIB0pYtW+LIkSNITU0tUhlV2HUSy8nJwaJFiwrt5aiqqoq1a9dKbefp6YnVq1cjJCQEffr0UdhT7N27d7h79y769+8Pb2/vQtNeVC9fvsTUqVPRokULAHnXdurUqTh58mSRgyIvXrzAn3/+WSrzXNSuXRt+fn5Sy44cOYJnz57B0tLyk1yHEydOICUlpdjDX718+RK5ubm4cOECcnJy0L17dxgbG+PGjRs4cuQIXr9+jcmTJxd4jKLmCWtra9SqVQunTp1C/fr1JQ1XClJefsOlpKTA19dXZmhWFxcX7N+/HwkJCVIBhJCQEBgZGcn06FCksPKuuG7evIkbN26gY8eOkt9JAFC1alVs2rQJRkZGUp9t3759qFGjBmbPni35DnBwcJA7FFtRr11xyyoiIqKvQfmYiY+IiCokcaVlRkYGAODff/9FSkoKnJ2dkZKSguTkZMlf06ZN8f79e9y7dw9A3lwlZmZmOH/+vNQxz507B21tbckLclpaGm7evIlWrVpBEASpY4qHzPn3338VpvHff/9FVlYWOnXqJFWJUKVKFbRp0wbR0dFISEiQ2qdRo0aSgAiQFyjo0KEDkpOTERkZKbVtnTp1pFpeVq1aFVpaWqhcubKk8h0AdHV1YW5uLjUU0aVLl5CdnQ0nJyepz5WamoqmTZvi7du3ePr0qdT5WrZsKVUJpq+vDwsLi1Ib4qhjx45SlbhNmjSBhYUFrl+/LlmmpqYmqSzJyspCSkoKMjIyYGdnh9jYWLx7907muJ06dfqodOXm5uLSpUuoWbOmVKWxWFGGHrGwsJC6J8rKyujYsSOys7MRHh4uWS6uyM/NzcW7d++QnJwsGULl4cOHMsf92HsibqEufjbK0rt377BlyxZ4eHgU2qJYkWPHjmHixImIiooqdFtvb29MmTIF9erVw6NHjxAYGIjff/8dw4cPx4EDB6S2zR9gyczMREpKCoC8skTeuZSUlNChQwfJv1VUVFC9enUIgiATXKlduzaSkpKQnp4OIK83VExMDNq0aYP09HSp51Och27dulXgZ1NRUZE8S9nZ2UhNTUVKSgoaNGiA9PR0PH/+HEDR83Z4eDjS0tLg7e0tFRD5cLvQ0FBUqlQJ9evXl0q3iooKatWqJTfdH05ob2VlhcaNGyM8PBzZ2dkFfk7xOW1tbWFqaip1TkNDQ5iZmUnOmZubi+vXr8POzk6qZ5eKispHlxFA8e6blpYWAODy5cvIycn56HMXR4cOHYo07KNIJJJsl5OTg9TUVCQnJ6N+/foQBAGPHj1SuK+amhpUVFRw584dJCUllVraK1euLAmIiNnZ2Uk9P4Vp2rRpqU38LQ565k8LAEn5W9rX4f79+9i+fTvq1q0rVb4Uhfi3UkpKCiZNmoSuXbvC3t4eo0ePhrOzMy5fvlzgPQU+Lk8Upjz9hpNXHoh7ggQHB0uWPX/+HA8ePICLi0uRhyD72PLuQ+JenR8Oxenm5iYT3BcHY8W9xMQsLS3RqFEjqW0/9vcvERHR1449RYiIqMyIK0DEL5fiSr4lS5Yo3Cd/pUTbtm2xZ88eSYvO5ORk/Pvvv3BxcZG08I2NjYUgCDhy5IjCyZ6Tk5MVni8uLg4A5A6JIF4WFxcn1erQ3NxcZlvxsri4OKkxxuUNd6CtrS13OBBtbW3JmM9AXmtZAJgxY4bC9H/42fK3OFR03I8hbzgQc3NzXLt2DYIgQCQSITs7G/v27cPZs2flnvfdu3eSCkexj60AS05Oxrt370pcUQ8o/mzAf/kEAKKiorBr1y7cu3cP79+/l9o+LS1N5hgfe098fHxw584d/PzzzzA0NISdnR2aN2+OVq1aSQWoPtbbt28VrhP3ktm9ezdyc3OlhpH61Fq2bImWLVsiOzsbT58+RXh4OA4fPoydO3eicuXKkhbH7969w+7du3Hp0iWZzyKvMszAwEBmjgPxPCUfPrfi/JqamgpNTU1JWbZt2zZs27ZNbroLq2AVl1snT56UlGP5ifNSUfN2bGwsAMDGxqbA7Z4/f46kpCSp3mcfys7Olro28p4NMzMzhIWF4c2bN3Lz+IfnzMrKUnhO8fBQycnJyMzMlHu+4g5FpCgdQNHum6OjI0JCQrBhwwbs2LEDdevWRePGjeHo6Cg36FSailMenj17FkFBQXj69Clyc3Ol1skrj8RUVFQwcOBAbNmyBcOHD4etrS0aNGiA1q1bFzrUYEEUlXfAf89PYUorIAJAZr6E/GkBSvc6REdHY9GiRTA2NsbkyZOlenFmZWXJbRAA5AVmtLS0JMEMQ0NDmXmQnJ2dERISgrt37xaarpLmicKUl99wenp6Mr8hgLyeWHZ2dggJCZH0IgkODoZIJIKzs3ORP+fHlncfiouLg5qamsz3ioqKCkxMTKTuifh3gbzfmeI0iH3s718iIqKvHYMiRERUZsS9GMQvd+JKvyFDhsh94QOkgxPiF+rz58+jT58+kiEl8g+7IH7hb9++PVq3bi33mPknv/7cFE22XNgkzMB/n23SpElyKwAAwNbWttjH/dQ2b96MEydOwMnJCX369IGuri6UlJQQHByM8+fPy1TSAChwGKXyJD4+HnPmzIGuri769OkDMzMzqKmpITc3FwsXLpQ7TNfH3hNzc3MsWbIE//77L27duoXbt2/j/PnzsLa2xty5c4tUyVgUw4cPV7huz549ePHiBU6cOIGePXtKWqMC/1UuJiQkQFlZudAKTV9f3yIPo5OfiooKqlWrhmrVqqFVq1aYOHEizpw5IykPli1bhps3b8LDwwN16tSBtrY2RCIRDhw4gNu3b8scr6D7omid+P6K/9u9e3e5Ey0DspWxHzp06BC2b9+Opk2bwsfHB5UqVYKysjLCwsJw+PDhIg35VhKCIKBKlSoYOXKkwm1KuxwRBAE1atRAnz595K4vbJ6E0kwHULT7pqqqip9//hkPHjzAzZs3cefOHWzevBl79+7FjBkzZMre0lTU8vDixYtYuXIlateujWHDhsHQ0BAqKip4/PgxduzYUWge8vT0RIsWLXD9+nXcvXsXp0+fRmBgIHr16iUzNFFRFZR3ipqni9JLJj953ylFSY9YaVyHmJgYzJ8/H9ra2pg5c6bMXA0XLlzA6tWr5e4rnpRb3KNQ3m8WceCwsKDGx+aJgpSX33AFPR+urq5YsWIFIiIiULt2bZw9exb16tX7bJOJF5QXP9W5yuvvXyIiorLGoAgREZWJd+/e4erVq9DU1JRMVC6uLNXS0pJpBSmPiYkJatWqJXmhPnfuHIyMjFC3bl3JNqamphCJRBAEoUjH/JD4RfnZs2eoWbOm1Dpxq8gPX6bFPTjkLSvNF28zMzMAeS+1H072XlaeP38u02L9xYsXMDIykhqep169ehg7dqzUdmfOnCnWuYo61AWQ13JUW1sbT548KdY58hPf7/w+vK9Xr15FZmYmfvjhB6lKVXl5ojSpqamhefPmaN68OYC8Mes3bNiAc+fOyQz1VFIF9UgCgDdv3iA3Nxd79uyRO7n6ggULoKSkhN27d5dKegpibm4uGVYM+G8YEWdnZ6lJZAF8kongxc+mqqpqicodADh//jyMjY2lxuYHIBPAKWreFqcpOjpaat6OD5mamuL27duws7OT6SmjyPPnzyV5Tyw2NhaqqqqSytqCnldTU1OkpKQUeq309PSgrq4u91mUt0wRRWkp7n0TiUSSeWwA4MmTJ5g+fToOHDiAiRMnFniu4qatJM6fPw9VVVXMnDlTKpCQv2dbYSpXrgx3d3e4u7sjKysLCxYswN69e9GlSxepIYTKAx0dHblBgdLoCfkx1+HFixeYN2+eJJAmbx6XRo0aKSxj8wdDKleuLCnX8nv9+jUASM1vI09p5AlFyuNvuA+1atUKf/31F86cOYPMzEy8efMG/fv3L9YxilLeiSdI/zA/yrvOxsbGCA8Pl5nrJDs7G69evZKabF3cC+XFixeoUaOGTBryK+61K82yh4iI6EtQ9s1FiYiownn37h2WLFmCtLQ0dO3aVVKh0LhxY+jq6mL//v1yxxZPTk6WaWXXtm1bxMXFISQkBA8ePICjo6PUi52enp5kYt0P59cAgPfv3yscsgLIm5RXTU0NR44cQVZWlmT569evce7cOdja2soMeRAeHi51royMDJw6dQq6uroFVkYWV+vWraGiooK///5b7jjWHzP+uYaGRomG0Th69KjU+PphYWF4/vy51CSoSkpKMvcxNjYWV65cKXYagaIN96GkpIRWrVrhwYMHcs9TlNaxz58/lxqaIicnB0ePHoWysrJkLG9xq+MPP19gYGChxy+Monsib/gL8RBJ4nkzSkPDhg0V/gF58+FMmjRJ5s/e3h4A4OfnJ6koLg2ZmZm4e/eu3HX37t1DamqqZJgTRfclIiICDx48KLU0idna2sLCwgJHjx7FmzdvZNanp6dLlSfyiNOcP2+mpKTIBA+LmrcbNWoEbW1tBAYGSnrvyNuubdu2yMrKgr+/v9x0yStXTp06JZnvAMjrBXjz5k00bNhQElgp6Hlt27YtXr16hdOnT8s9pziPKykpoWnTprh7967U3AfZ2dkKh4eRR1FainPf5D1bFhYWUFNTk7q+6urqePfuXZFbiBenXCuMkpISRCKR1LmzsrJw9OjRQvfbmOq3AAA/O0lEQVTNzMxEZmam1DI1NTVYWFhAEIRSSV9pMzU1xfPnz6WGx0tMTJTM2VASH3sdXr16hblz50IQBMycOVNhwwgDAwOF5Wv+3hWOjo54+/atzLN+4sQJiESiQiu/i5MnxHlRXnkhT3n8DfchNTU1ODo64tKlSzh+/Dg0NTUl31FFVZTyThwgunPnjtS+x44dkzmeOMBy6NAhqeUnT56UuY4NGjSAqqoqjh07JvW779mzZ1JzmwHFv3bFLauIiIi+dOwpQkREn1RUVJTkxfrdu3d48uQJLl26hLS0NHh4eKB79+6SbTU0NDBq1Cj8+eefmDhxIlxcXGBkZISkpCRER0fj2rVr2Lx5s1TrRgcHB2zevBl//fUXAMDJyUkmDd9++y1mzpyJH3/8ES4uLrCyskJWVhZevHiBy5cvY9y4cWjcuLHc9Ovo6KBPnz7YunUrZs6cibZt2yIzMxMnTpxAbm4uBg8eLLNP1apVMWfOHHh6ekJTUxMhISF49eoVRo4cWeSW10VRpUoV+Pn5YePGjZgyZQratGkDAwMDvHnzBg8ePMCTJ0+wZs2aEh27Zs2auHv3Lg4cOAAjIyOoq6vLtIyUJyMjA3PnzoW9vT0SExNx9OhRVKpUSeo+t2jRAv/88w+WL18OOzs7JCQk4Pjx47C0tMTjx4+LnEZLS0toaGjg9OnT0NbWhra2NoyNjWV69Ij169cPt2/fxp9//gknJyfUqFEDGRkZiIiIQKNGjWQm3P1Q1apVsXTpUri7u8PQ0BCXLl3C/fv30bNnT8kQFI0bN4aqqipWrVoFT09PqKur4/r166Uybreie7JgwQLo6OigTp06qFy5MpKSknDq1CmoqqoWqbLn2LFjSEtLkwSznj59in379gEA6tatq3AYoQ/p6enJPZ+4MqZRo0alMu+DWGZmJmbPni2Z5NbU1BS5ubl4+vQpzp49C3V1dcnQNpqamqhfvz7OnTsHDQ0N2NjY4Pnz5/jnn39gaWmJmJiYUksXkNfidsyYMZg3bx4mTZoEV1dXmJmZIS0tDc+ePcPly5fxyy+/FHg9WrRogb///huLFi1CixYtkJSUhNOnT8PQ0FAmMFGUvK2hoYFvv/0Wy5cvx9SpU+Hi4oLKlSsjISEBly5dwoIFC6CtrQ0HBwdcv34dAQEBePDgARo1agRNTU3Ex8cjPDwcZmZmmDBhgtT5NTQ0MHPmTDg7OyM9PR3Hjh2Dmpoa+vbtK9mmoOe1S5cuCA8Px9q1a3Hz5k3UqVMHKioqiIuLw7Vr19CyZUtJa+4+ffrg5s2bmDdvHjw8PKCnp4cLFy4Ua4JjcQvrTZs2oXnz5lBWVkb9+vVRqVKlIt+3ffv24datW2jatCmMjY2RnZ2NCxcuID09XWr4n5o1a+LGjRvYuHEj6tSpAyUlJTRr1kxh74LilmsFadGiBS5fvox58+bByckJ6enpCAkJKdLwU7GxsZg9ezZatmyJqlWrQktLC48fP8Y///wDOzs7ufNelTV3d3dcvHgRCxYsQPv27ZGWloYTJ07A3Ny8xBOIf8x1SE9Px9y5c5GYmIiuXbsiKioKUVFRUtvI+81SEB8fH1y+fBnLli1Dx44dYWRkhLCwMNy4cQOdOnVSOGyVWHHyRPXq1SESiRAYGIisrCyoq6ujatWqsLKyknvs8vgbTh5XV1ecOHEC169fR4cOHYo9HFtRyzs7OzscP34cIpEI5ubmuH37ttxeS02aNEGTJk1w9OhRJCUloW7dunj69CkuXLgAExMTqSCFnp4eunXrhj179mD27NlwcHBAamoqjh8/Dmtrazx+/FgqqFSca1fcsoqIiOhLx6AIERF9UsHBwQgODoaSkhI0NTVhZGQER0dHuLi4yHT9ByCp5N2/fz9Onz6N1NRU6OnpwdLSEgMHDpQJKujq6qJx48a4fv06bG1t5U6IbmxsjF9//RUBAQG4ceMG/vnnH2hqasLY2BgeHh6FTkrapUsX6OvrIygoCLt27YKKigpq1aoFX19fuRVVLVq0gJGREQ4ePIi4uDgYGxtj9OjRcHFxKd7FKwJ3d3eYm5vj0KFDOHz4MDIzM1GpUiVYW1sXe0iI/IYOHYoNGzYgICAAGRkZMDIyKlJQZOzYsTh69Cj8/f2RmZmJunXrYvDgwVLjVvv5+UFNTQ2XL1/G5cuXYWFhgW+//RbPnj0rVlBETU0N48ePx99//43NmzcjOzsbzs7OCisP9fT0sHDhQvj7++Pq1as4d+4cdHR0UKNGjSL14KlWrRr69euHv//+G8+fP4e+vj6++eYbeHl5SbYxNTXF9OnTsWvXLvj7+0NNTQ1NmzbFuHHjCpy4uigU3RM3NzdcuHABJ06cQFpaGvT09FCrVi107969SEGIQ4cOSVXUPH78WHIfevbsWeSgyOemra2NkSNH4ubNm7hy5QrevHmDnJwcGBgYwN7eHt7e3lKff/z48di6dSsuXbqEkJAQWFtbY8qUKTh79mypB0WAvIp3cbkTGhqKpKQk6OjowMzMDD169Ci0UtnHxwc5OTkICQnBnTt3YGxsDG9vb2hoaMjMPVDUvO3g4AB9fX3s378fR48exfv371GlShU0atRIMg6/SCTCuHHjYGdnhzNnzmDv3r0A8lqy16lTB66urjJp/eabb3Dr1i0EBgYiLS0N1apVw8CBA6UqTwt6XlVUVPDTTz/hyJEjOHfuHG7cuAEVFRUYGhqiQYMGUkEGMzMzzJkzB1u2bMGhQ4egoaEBe3t7eHh4YMqUKUW6N23atMGjR49w8eJFXL58GYIgYNasWahUqVKR71vz5s2RmJiI8+fPIzk5GRoaGrC0tMSUKVPQsmVLybk6d+6M2NhYXLp0CadOnYIgCFi5cqXCisbilmsFcXJyQlpaGo4dO4atW7dCX18f7dq1Q61atTB//vwC961cuTKcnZ1x+/ZtXL58Gbm5uahSpQq6d+8uVeaVJ/Xq1cOwYcNw8OBBbN26FWZmZvDz80NMTEyJgyIfcx1SUlIkZaui3oLFDYro6Ohg7ty52LVrF4KDg5GWlgZTU1MMHjy40MC++HxFzRNGRkYYMWIEDh48iPXr1yMnJwc9e/ZUGBQByudvuA9Vq1YNNjY2iI6OLtHvsqKUd0De76ENGzbg9OnTUFFRQbNmzfDjjz9i6NChMsecNGkSdu7cidDQUFy7dg3Vq1fHTz/9hO3bt8sEUnr27AkNDQ0cO3YM27dvh6mpKfz8/CTf3fmDPMW5dsUtq4iIiL50IuFTzdJIRERUwcTFxWHs2LHo2bNniSaK/pLt2bMH/v7+WLly5WebsPRz8vX1lUx2S0R5goODsXr1asyaNQv16tUr6+QQERXJ9OnTkZmZiSVLlpR1UkrNr7/+itu3b2PLli2S4ReJiIhIMX5bEhEREREREdFX7+HDh3j06BHatWtX1kkpEXnzUcXExODmzZuoV68eAyJERERFxOGziIiIiIiIiOir9eDBAzx79gyHDh2Crq4uOnToUNZJKpEbN25g//79aNGiBfT19REbG4tTp05BSUmpwvVSJiIi+hgMihARERERERHRV+vEiRM4e/YszM3NMXHiRGhpaZV1kkrEwsICBgYGOH78OFJTU6GhoYE6deqgV69exZ5fhYiIqCLjnCJERERERERERERERFQhcMBJIiIiIiIiIiIiIiKqEBgUISIiIiIiIiIiIiKiCoFBESL6aqxatYoTDFK5FRwcDF9fX9y5c6fQbceMGYPZs2d/+kT9vzt37sDX1xfBwcGf7Zzlga+vL1atWlXWySiyr72MKw/3ozykgUrf7NmzMWbMmEK3q6hlIZU/n/t3wJ49e+Dr64u4uLhCty2LcrKoz/CXJC4uDr6+vtizZ89nPe+5c+cwceJE9OvXD76+vkhLS/us5yciIiovONE6EZV7d+7cwZw5czBs2DC4u7uXdXLoIxw+fBgaGhpo3759kbaPjo7GlStX4OTkBFNT00+atjt37uDOnTvw9PSEnp7eJz1XRRYXF4fg4GA0a9YM1atX/+TnK26eo9LB54mCg4ORnJyMrl27lnVSqAIore+WuLg4jB07VvJvkUgELS0tGBoaonr16mjTpg0aNmxYGkmmL9iXWr49f/4cq1at+r/27jsqqqNv4Ph3kQ4iooAgxQJGQLGDihRRkEQTS5TYosaYmGLypCePJUoUS0w0xfimGGNv2AsoFkREQRHBGkWNgqBCBKQ34f3Ds/fhsgvsIqbO55ycJMvdW2Zn5t475Te4urry7LPPoquri4GBwZ99Wn9pZWVlREdHk5CQQGpqKvn5+bRo0QJ3d3eef/55zM3NZdtv2bKFrVu3qt3X7NmzcXNzk/4/IyODFStWcO3aNZo3b86IESPw9fWVfaekpIR3332XwMBAhg8f3ujXJwiC8G8mOkUEQfjHmDp1Kq+88sqffRpCHcLDwzE3N9eqU2Tr1q106NDhD+kU2bp1K15eXqIR9wnKyspi69atmJub/yGdItrmOaFxNKQ8rVu3Dh0dMYn5n+Lo0aOkp6f/rRoNXVxcWLduHbq64hXp76ax7y3du3enb9++ABQXF5Oenk58fDxHjx6lZ8+evPPOO+jr6z/2cQTNzJw5k6qqqj/7NCR/x/oNHt2bKysrmTBhAm3atPmzT+dvISsrixUrVuDi4kJAQADNmjUjLS2NQ4cOER8fz8KFC7GwsFD53sSJE2natKnss9atW0v/XVlZyeLFiwEYP348v/76K8uXL8fGxoYOHTpI24WFhWFkZMSzzz77hK5QEATh30s88QuC8I8hGjEEQRD+vkQD4x+rpKQEQ0PDP/s0/lJ0dHT+kvmwsrKS8vJyMaL7D2Rvb4+Pj4/ssxdffJEff/yR6OhoVqxYwRtvvPEnnd2/Q1VVFaWlpRgaGopnfDWKi4sxMjLS6jt5eXkAGBsb/+nn8ndhZmbGokWLVDqR3N3dWbhwIXv27GHixIkq3+vVqxdWVla17vfOnTukp6czZ84cXF1dCQgI4OrVqyQkJEidIqmpqYSHhzNr1ixRBgRBEJ4AUbMKgvCP8d133xEdHS2LzTtnzhyysrIICQlh1apVnD9/HoAePXowZcoUlZeC33//nbCwMJKSksjLy8PCwgIvLy9Gjhwpayi5desWW7ZsISUlhcLCQkxNTWnTpg2jR4+mbdu29Z5rYmIie/bs4caNG1RWVmJpaYmHhwejR4+WtsnNzWXTpk0kJiZSUFBAixYt8Pb2ZsSIEbIHY+V1r1y5kjVr1nD69GkqKipwc3NjypQptGzZUtpWOaV7yZIlHDlyhJiYGIqKinB2dmby5Mk4ODjIzrO0tJRdu3YRGxtLVlYWRkZGdOnShbFjx8r2C5Cfn8/27dtJSEjg/v37mJiY0L59e4KDg2nXrp20FkJWVpZsXYTaYilXn34+f/586fORI0dK38/Pz2fbtm2cPn2a7OxszMzM6NWrF6NHj8bU1FT6TmZmJps3b+bSpUs8ePAAExMT7OzsGD58OO7u7lIaArz77rvS99544w38/PwAzfMGQFxcHFu3buXOnTtYWFgwaNAgTExM1F5nXa5evcratWv57bffMDExwdfXl+DgYNnv/+abb2JpaakSe1yZfsuWLZO9lF28eJH169dz69YtTE1N8fPzk03lr+7u3busXr2aCxcuoKurS7du3Zg4cSKvvPIKvr6+KvG9jx49yoEDB7h9+zYKhQInJyeCg4Pp2LGj9Pfly5cDsGLFClasWAEg7au0tJTt27dz8uRJ7t+/j76+PlZWVvj5+fH000/Lzguod/aQpnnu0qVLrF+/nps3b2JqasrAgQMZOXIkCoVCtt2VK1fYvn07V69epaysDFtbW55++mn8/f2BRw2Xr7/+OnZ2dsyaNUvlfL766ivOnDnDjz/+WGfjQWVlJdu3b+fw4cPk5eVhb28vqxtqOnfuHNu2bePGjRsAtGvXjuHDh9O1a9c60wf+95vMmjWLlJQUDh06xIMHD7Czs2PixIm4urrKti8qKiIsLIz4+HhycnIwNzfH09OT4OBgqT7VpDypExwcLMtXyjA2I0eOxMbGhu3bt5OZmYmtrS0TJ06kU6dOnDt3jo0bN5Kamkrz5s0ZM2YMXl5e0j6r78PS0pJdu3aRlZWFlZUVw4cPV2n0rK6goICpU6fi7e3Na6+9pvL3GTNmkJOTw7Jly+qc4VJSUsKGDRs4ceIEpaWltGvXjgkTJrB27VqysrJksflrpoGSuvtbUlISR44c4dq1a+Tm5mJkZISbmxtjxozBxsZGbdp6eXkRFhbGrVu36NOnD2+++SYpKSkcOHCAK1eukJ2djb6+Ph06dCA4OFhlxL2yvpk8eTKrV6/m6tWr6Ovr4+XlxYQJE6S66c033yQrK0s6tlL1+qi+8lTdmTNnZPdMR0dHhg8fTs+ePWXblZaWsnHjRmJjYykuLqZ9+/a8+OKLtf42NSnDdVbPq8oyMnPmTC5dukRUVBRFRUV07NiRqVOn0rJlSyIiIggPDyc7OxtHR0deffVVWQOach8zZszg/PnzxMTEUFBQQPv27ZkwYQJOTk4q5/D6669TVFREZGQkmZmZTJ06FT8/P43KYHh4OKtWrSIkJAQXFxfZNd67d4+33nqL5557jvHjx8vOsa76u2b6FBYWEhERQU5ODu3atWPKlCk4OjoSGxvL9u3buXv3LtbW1kyePJlOnTrJzqGyspLw8HCOHj3KnTt30NPTw9XVlTFjxmBvb6+SbvXVT/XdWxqLnp4eU6dO5erVq0RHR/P8889jbW0NaFceT506xZ49e7h9+zbl5eWYm5vz1FNP8fLLL6s8l6amptZZ1pTS0tIICwvj4sWLFBcXY2lpib+/P88++6xK/XTgwAH27dvH/fv3sba2ZtiwYQ1KD03vnbGxsezdu5e0tDSaNGnCU089RXBwsCzfV6+rW7Vqxe7du8nIyGDYsGEEBwdLz/PK+rKuEEXKvytdv36dLVu2cOXKFcrLy7G3t2fw4MF4e3vLvtNY9VtsbCwxMTHcvHmTBw8eYGZmJj071wy1pCnlPeDHH39kzZo1JCcnk5+fL12nJs+o1c9VGSKuehnR9DlX+VvMnDmTNWvWcPnyZUxMTKTfRtO8qGl6K6WlpbFt2zYuXrxIYWEh5ubmdOrUifHjx8tmpGpSl4Hmz5JNmzZVmfEB0K1bNwwMDLh9+3at3y0qKsLQ0FDtM0JZWRmA9H6gUCgwMTGRPq+qqmLFihV4e3urPIsJgiAIjUN0igiC8I9XUlJCSEgIbm5ujBs3jpSUFKKjo2nSpInsZfnevXvMnDmTJk2aMGDAACwsLLh27Rq7d+/mt99+Y/r06SgUCvLy8vjss88wNjbm6aefplmzZuTm5nL58mXS0tLq7RTZu3cva9aswdbWliFDhmBubk5GRganT5+WGj4LCgqYMWMG2dnZBAQE0Lp1ay5cuMDWrVu5ceMGH3/8scpLZ2hoKJaWlowZM4aMjAz279/PsmXL1C7U+d1332FiYsLzzz9Pbm4ue/fuZfHixXz11Vc0adIEgIqKCubNm8fNmzfp378/Dg4O/P777xw4cIBLly6xaNEimjVrBjwaeTZjxgwyMzPx9fWlffv2FBcXc/nyZa5cuUK7du2YNm0aq1evxtjYmFGjRtX7u3l6epKbm8uhQ4d47rnnpA4bR0dHKY1mzpxJXl4eAwcOxNramvT0dCIjI7ly5QqhoaHo6+tTUVFBaGgoRUVFBAQE0LJlS/Lz87l27Ro3btzA3d2dgIAAiouLOXXqFOPHj5deWp966imt8gbAyZMn+eqrr7C1teWFF16goqKCffv2aR2S6/79+yxYsIB+/frh5eVFUlISO3fuJDc3t8GjU5XpYmZmxogRI9DT0yM6OpqkpCSVbfPy8pg9ezYFBQUMGjQIKysrkpKSZB1U1a1Zs4Z9+/bh6emJr68vZWVlREVFERISwsyZM3Fzc8PFxYXhw4ezY8cO/Pz8pIYy5QvpihUrOH78OAEBATg6OlJaWsrt27e5dOmSrFNk7ty5APUu9KpJnrt58yZLlixhwIAB+Pj4cOLECcLCwrC0tJQ14J86dYqlS5fStm1bhg8fjr6+PmfOnOH7778nNzeXESNGoKOjg6+vL7t27eL333+XdRwWFhZy+vRp+vTpU+9oyl9++YUDBw7QpUsXevToQVZWFkuWLMHS0lJl27i4OJYuXYqVlRUjRoxAoVAQFRXFggUL+M9//iOFf6nPhg0bAHjmmWcoLy9nz549fP7551JdAY/qhLlz53L9+nV8fX1xcnLixo0bREREcPnyZebNm4eenl695UlbZ86cIT8/n4CAAHR1ddm9ezeLFi3izTff5OeffyYgIAAfHx/279/Pt99+S7t27VQaIZUdp4MGDcLY2Jjo6GiWLVtGZWVlrR01pqam9OrVi5MnT/LSSy/JRumnp6eTkpLCyJEj6+wQqaqq4ssvvyQ5OZm+ffvi4uJCWloa8+bNU9vIoo1jx45RUlLCgAEDaNasGRkZGRw+fJjLly/z5ZdfqtQ5169f59SpUwQEBNC/f3+p8fX06dNkZmbi7e2NhYUF9+/f59ChQ8yePZtFixbJwn0AZGdnM2/ePPr27YunpydJSUns378fU1NTqdFt4sSJbNy4kZycHCZPnix9V3lOmpQnpYiICH755Rfc3d154YUXUCgUxMbGsnjxYqZNmyZr2Fy6dCmJiYn06dMHV1dXbt26RWhoqKyTvKHWr1+Pvr4+Q4cOJScnR7pv9unTh5iYGAYNGiQNJPjiiy/45ptvVPLGunXrAHj22WcpKiriwIEDfPbZZ8yfPx87OzvZtuHh4RQXF+Pv74+pqSm2trYal0EvLy/Wrl1LTEyMSqdITEwMgCzdNKm/q4uIiKCiooKgoCDpmufPn8+YMWMICwuTyuquXbtYvHgxy5cvlw0M+Oqrrzh9+jTe3t4EBgZSUFDAgQMHmDlzJvPnz1fJc/XVT/XdWxqTrq4u/fr1IywsjOTkZGmtO03L44ULF/jyyy/p2LEjo0aNQl9fn99//53ExEQKCwtlnSKalDV4NIBi7ty5tGjRgiFDhmBiYiINgLh79y5Tp06Vtt29ezfr1q3D2dmZoKAg8vPz+eWXX+oc0a6OpvdO5TNv27ZtGT16NCUlJRw8eJDZs2cza9YslYbqU6dOkZOTw6BBg7CwsKBFixZqj+/p6any+xYVFbF27VpZGl69epWQkBCMjY0ZPHgwRkZGxMbG8u2335KTk6MS/qox6rfIyEjMzMwICgrC1NSU3377jaioKFJSUvj888/R09PTKq2rCw0NpWXLlgQHB5Ofnw9o/ow6bdo0Tp06xalTp6TQTso01OY5F/73btWpUyfGjRtHaWmplN6a5kVN0xvg119/JTQ0VDo/GxsbsrOzOXPmjDQoCrSryzR9lqxNYWEhZWVltT7ff/jhhxQXF9OkSRNcXV0ZO3asbKCBra0tJiYm7Nixg7Fjx3LlyhVu3rwp5UlliLYPPvigQecnCIIg1E90igiC8I+Xn5/P888/zzPPPANAYGAghYWFxMTEMHnyZKmBcuXKlejq6vL5559LjVUDBw6kTZs2rFy5krNnz9K9e3euXLlCfn4+//3vf2Wj3DSRlZXF+vXrcXZ25tNPP5U1slWPlbxz506ysrJ46623pIaLQYMGsXLlSvbv309CQgK9evWS7fupp55i0qRJss/Cw8O5ffu2SmNLixYtZA/ZTZs2ZfXq1Zw7d45u3bpJ301JSeGzzz6Txbbt3bs3n3zyCbt375ZG4G7YsIF79+7x/vvv4+npKW07bNgw6bp8fHzYvHkzZmZmdY7OVnJ0dKRDhw4cOnSITp06qYx637RpE7m5uSxatEj2YtypUycWLVrEkSNHCAoK4vbt29y5c4d33nmn1gbiDh064ODgwKlTp+jRo4dKg4ymeaOyspLVq1djYWFBaGio9GLu5+cnGzGviXv37jFlyhSpwSUoKIilS5dy9OhRnn76aY1mJNW0Zs0amjRpQmhoqNTQEBAQoPaFa9euXeTk5PDRRx9Jo7GDgoL4+uuv+e2332TbXrt2jb179zJu3DiGDh0qfR4YGMj777/PunXrWLBgAdbW1ri7u7Njxw6cnJxU8kFCQgIDBgyQNTI8Dk3yXGpqKgsWLKBdu3YA+Pv788YbbxAZGSk17JSVlfHDDz/QpUsXWYdkUFAQX375Jdu2bSMgIICmTZvSv39/du7cKY0iVoqNjaW8vJz+/fvXec63b98mMjKSnj178uGHH0rHcnFx4fPPP5dt+/DhQ1atWoWZmRnz58+X8qbyN/3ll1/w8PDQOOTCvHnzpE7R1q1b88UXXxAbGyvlwSNHjnD9+nVGjx4ta7S2s7Nj7dq1HDx4kGeeeabe8qStu3fv8tVXX9G8eXPp3BYuXMjXX3/NokWLpA7TTp068f7773P48GHZCHh4NML0888/l7YdOHAgH374IevWraNfv361ppG/vz8nTpwgPj5eloeOHj2KQqGoc+YLwNmzZ0lOTmbw4MGy8BqOjo789NNPaju6NDV16lSVcEo9e/Zkzpw5HD58WGVB1tu3b0thOqp7/vnnVfbj7e3NBx98wL59+3j11Vdlf7t79y4ffvihdA8KDAzkww8/5ODBg1IjloeHB+Hh4RQUFKiUPW3KU3Z2NmvXriUwMJApU6ZI+wgKCmLmzJmsXbsWLy8vdHR0SEpKIjExkaeffpqXXnpJ2tbe3p5ffvnlsdIaHjWGh4SESB0dlZWV7Nmzh8LCQpYsWSKNpDYxMWHlypUkJydL91OlkpISFi9eLIUt6927Nx9++CEbNmzgo48+km2bnZ3N119/LevQiYyM1KgMNmvWDHd3d+Li4pg8ebIsf8fExODg4CANMNC0/q6uoKCAL7/8UroO5TWvWrWKr776SuoItbW1ZeHChbJ65OTJk8TFxanck5X3yc2bN/Pee++ppH9d9VN995bGpky7O3fuSJ9pWh4TEhIwNDTk008/la4HUDsbUJOyVlVVxQ8//ICtra3UKabctkWLFuzdu5dnnnkGe3t7CgoK2LJlC+3btyckJETKF56ennz88cdapYEm986CggI2bdqEo6Mjc+fOlcqI8rdeuXKlyn0tIyODJUuW1Nuh5ejoKP0O8Oh+uHDhQgDZM82qVauARw3gyn0OGjSIWbNmsXnzZnx9faUBPvD49RvA9OnTVfKCs7Mzy5cvJz4+nn79+tV5bXVp164dr7/+uuwzTZ9RfXx8uHv3LqdOnVIJ7aTpPpTy8/MJDg5m5MiR0mfa5EUlTdK7srKS77//Hh0dHT7//HPZeQcHB1NZWQk0rC57HNu3b6eqqkplxpGJiQkBAQF06NABIyMjbt26xd69e/n000+ZPXu29E5lYGDA1KlTWb58OSdOnACgX79+9O3bl4KCAtatW8e4cePEOoeCIAhPkFjNUhCEfzwdHR0CAgJkn7m6ulJZWSlNfy8sLCQpKQlPT0+qqqrIy8uT/nF3dwcehaiB/01zPn36tDTFWVPx8fE8fPiQUaNGqbwwVR+BdebMGSwtLVVenJQhDk6fPq2y76CgIJVrhP9ND69u0KBB9W4bGxtL27ZtadWqlSw9LCwssLGxkUKRVVZWEhcXh7Ozs6xDRN11NZaqqipOnDhB586dMTY2lp2fs7MzBgYG0vkpOyaSkpIoKirS+lja5I3r16+TnZ3NgAEDZCMVlSEItGFsbKwSRka5yOKZM2e0vo7c3FxSUlLo06ePbOSlkZGRSvlQHsPa2lolPM3gwYNVto2NjaVJkyb07dtXlj5lZWV07tyZ69evU1hYWO85Ghsbk5KSQmZmZp3bfffddw0e2VfTU089JTXqwKPwKM7OzrKycO7cOfLz8/H19SU/P192jd27d6e8vJxff/0VABsbG1xcXDh69KjsONHR0VhZWdUbAiEhIYGqqiqGDBkiKzs9e/ZU6Vy4ceOGNJus+qwD5Qv5gwcPuHr1qkbpEBAQIGugU1cnJCQkoKenp5IHgoKCMDIyUlsvNYZevXpJHSLwvxknys4XJXt7e4yNjdXWeV26dJFta2hoyMCBA8nLy6szjTp37oylpaXs96ysrOTYsWO4urrWO7o6ISEBUC031WdqNJTyHlJVVUVRURF5eXnY2dlhYmLC9evXVbZv27at2vxX/V5UUlJCfn6+NDNB3X5atGih0inv6urKgwcPKC4urve8tSlPcXFxVFRU4OPjI9uuoKCA7t27k5ubS2pqKvC/+2LNxWgDAgIaJdb9gAEDZDM/lPnQx8dHFlpGOfJdXT4cOHCgbB0XBwcHunbtSnJyMhUVFbJtfXx8VGa4aFMGvb29KSgokM0EvHbtGnfu3JE14jWk/vbx8ZFdh/Kae/bsKQsPpEyjms8WzZo1o1OnTrLj6erq0qFDB+neXZ0m9dMfSXnt1fO7puXRxMSE0tJSEhMT6104XJOylpqaSlpaGv369aO4uFiWpspOOWWanjt3jrKyMoKCgmQdZW3atJGeZzSl6b2zrKyMZ555RlZGWrZsSb9+/bh58ya///67bL/du3dv0AyfX375heTkZKZOnSrlu9zcXK5du0afPn1k+9TT02PIkCGUl5erzJR93PoN/pcXKisrpbygzLPq6lRtKAd4KWnzjFqbhu6j5rlokxeVNEnvmzdvkpGRQWBgoNp7rrJe1rYue5xnybi4OPbt20e/fv1UBm0NHjxYCjPr4eHBqFGjmDdvHlVVVaxevVq2be/evfn+++8JDQ3lu+++4+2330ahULB+/XpsbW3p378/qampzJ07l1dffZWQkBDpnicIgiA8PjFTRBCEf7zmzZurTFVXNjQUFBQAj0b7VVVVER4eTnh4uNr9KBcndHFxoV+/fuzYsYN9+/bRoUMHunbtipeXV63T/JWUowqrj25TJzMzEzc3N5UOBQsLC4yNjaXOnOpqvigoO2+U11hdzRGz6rZNT0+nrKxMNjq3OmUjZV5eHkVFRfVeU2NSNoopwwDUtg08SpehQ4eye/duYmJicHJyokuXLvTt2xdbW9t6j6VN3lA25qvbb83ZOvWxtrZWGb2u3G99nQbqKL+jbtS+us+ysrJU4sBXP4fqMjIyePjwYZ2x2/Py8updV2XChAksW7aMadOmYW9vj5ubG717936isZTVjR43MTFRKQvwKDRPbR48eCD9t7+/P9999x2XL1/GxcVFCrUUHBxcbydhXb+Tra2tdC7Vt1WXt5SfZWZmapR+NdOhZh2p3JeVlZVKh66enh5WVlYNypeaqLl+kTIf1fxc+Td1HXDq8m318lRbGikUCvr3709YWJgUEu3cuXPk5OQwduzYes89KysLfX19lXPV1dXFyspKo87C2mRkZLBhwwbOnTtHSUmJ7G/q9qtc+6CmnJwc1q9fz5kzZ1S+p64BqrYyA4/yS30dENqUp4yMDABmzpxZ67bV69/a0tra2vqx0hpqv2/WPJ6ys0vdvVddubaxseHs2bPk5OTIjqGuYVibMtirVy8MDQ2JiYmROrePHz+OQqGQDbhoSP1dMy2U11xbWa1Znz548KDWZwt4FKqv+v1Pk/rpj6Qsb9XzuqblcdCgQZw6dYrFixfTtGlTXF1d6datG3379pV1NIFmZU1ZntauXcvatWvVnq+yPNX3HJCcnFz3hWtwbjXvGVD/Pap6vmlIh0hERASRkZEMGzZMNnND0+NX97j1GzzqfNy4cSO//vor5eXlsr89bj1UM320eUatTUP2YWZmptKxr01eVNIkvZUdbfW9ZzTWs2h9zp07x7fffkvHjh3Vrjemjr29PT179iQ+Pp7CwkLZORgbG+Ps7Cz9v3LNooULF1JSUsLcuXPp3r0748aNIzIykrlz5/LNN9/8Yxe2FwRB+COJThFBEP7x6ov3DkhTrwcMGECfPn3Ubqsc/ahQKHj77bd57rnnOHv2LJcuXWLz5s2EhYXx/vvva7S48ZNQ13Vqum31UYtVVVU4OTnVusDz48REflzK8+zRo4dsrYnqqr+sjRs3Dn9/f86cOcOlS5fYvXs327Zt45VXXlG7qG912uSNvxLlef9Rx9LT06sz/IaFhUW9+/H09MTFxYXExEQuXrxIfHw8+/fvp3///irhIhqLJuVGmd8mT55ca0da9UaX3r17s3LlSo4ePSrNGlEoFPj6+jbOST8BmtQJf5bazu2POmc/Pz/CwsI4evQoI0eO5OjRoxgZGdG7d+9GPU5dapbn4uJi5syZQ0VFBcOGDcPOzk5qUP3666/VpkHNhnTlfkNDQ7lz5w5Dhgyhbdu2GBkZoVAoWL16tUrjLmh2T62LNuVJed3vvfderTNrGhJKsCH+6Hyo7vfS9vseHh7ExcVRXFyMgYEBsbGxuLq6ygZwNKT+fpy0qKqqomXLlnU2Jtbcz1+tfrp16xaAtHaRNuXRzMyMhQsXcunSJZKTk7l48SLff/89W7duldZiUNKkrCn/PWLEiFo7d7VdL0QT2jxzaqP6jBJNJCUlsXr1ajw8PBgzZsxjH/9x67esrCxCQkJo2rQpo0ePxsbGBn19fSorK5k/f36j1wuN8YzakH2oq58akhcfN72ra6xn0bpcvHiRxYsX4+DgwMcff6xVfrWysqKqqoqCgoJaO2YqKytZsWIFzzzzDA4ODhw/fpzCwkImT56MgYEBL730EseOHePMmTOPFYZNEARBeER0igiCIPBo5JVCoaCqqkrjEAJt2rShTZs2DB8+nPv37/PRRx+xZcuWOjtFlC/Qt27dkoWDqcnKyor09HSqqqpkI8tzcnIoKip6Ii+4NbVq1Yr8/Px608PMzAwTExOpkeCPoByhVlZWpvHvZWNjw5AhQxgyZAiFhYVMnz6djRs3Sp0itY3g1yZvKH8X5ejm6m7fvq3ReSrdu3dPZbSscr/Vf39TU1O1Iw9rziZSfqf6TAMldZ9ZWlqqvQ51n9nY2JCcnEzr1q3rnS1V30wJMzMz/Pz88PPzo7Kykm+//ZaoqCieffZZrWfbNBblyExjY2ON8puBgQFeXl7ExsYyadIkjh07RqdOnTRa06D671Q9zjmopr1yW3V5S/mb1jY7oCGsrKy4dOkSpaWlsgaR8vJyMjMzZQuIPomweY+jrrxcX33asmVL3N3diY6OJigoSFogWpNGa0tLS5KTk6VZJkoVFRVkZmaqNIzUNtOlZnm+cOECubm5vPHGG7J1TcrKyrQaPZ+amkpqaiqjRo1i1KhRsr8VFBRovB6NOnXVqaBZeVLeM83NzVUWZK7Jysqq1rS+d+9eoyy2/rjS09NVQhLeuXMHPT29Op8JlLQpg/AohNaxY8eIj4/H3NycBw8eqDQca1N/N4ZWrVpx4cIFXF1dHyt/1fRH1TkVFRXSjBvl85625bFJkyZ07tyZzp07A4/WHlqwYAEREREq6yHVR1lG9PT0NH5GSU9PV1kTT91zwOOqfo+qPhK++vEe53n29u3bLF26FAcHB6ZNm6aSB+q6Ryo/a+g9srb8dvr0aUpLS/nvf/8r6xhQdw9qDA15f3kS+wDt8qI2lPeMW7du1dkR8KTrsl9//ZVFixZhbW3NjBkztA6BeffuXRQKRZ1rhERERFBQUCCt1XL//n1MTU2l+t7AwABTU1Pu37/f8AsRBEEQJGJNEUEQBB41xLq7uxMTE6M2Vmt5ebm0HkVBQYHK6KUWLVpgZmZGfn5+ncfx9PSkSZMmhIWFqaxHUn2fPXv2JCsrS1p4T2nnzp0AKvF3nwRvb2/u3bvH4cOH1f5dOZVeR0cHT09PUlJS1Iayqn5dhoaGWjXYKUda1vyOjo4Offv25cKFC2rjHFdWVkq/RVFRkUqsdhMTE6ysrGS/pfJYNRsktckb7du3p3nz5hw+fFi2fkl2djaxsbEaX7fyvI8cOSL7bO/evQCyxS5btWpFeno6ubm5suPVXN/B3NwcJycnTp48KXuZKi4u5uDBgyrH7969O/fu3ZPWRFDat2+fyrbKl9SNGzeqHdlXPVxCbelcWVmp8pmOjo4ULqGhYVK0zXPqdO3alaZNm7Jjxw61McXz8vJURvL7+/tTXFzMTz/9RE5OTr0LcispG0z37dsnS8uEhASVRqt27dphYWHBoUOHZNdYVFREZGQkzZo1U2mIehw9e/akrKyMiIgI2ecHDhyguLhY1thb2+/8Z0lOTpaV35KSEg4dOkTTpk2lRU/r4u/vz71791i5ciXl5eUa/549evQAVMtNVFSU2jWObGxsSElJkYVcuXHjBleuXJFtpxxdW7O87dmzR6vRtbXt59ixY+Tk5Gi8H3UMDAwoKipSKRvalKc+ffqgq6vL5s2bVepxkNctyvy3Z88e2TYHDx7UeC2AJ+3QoUOy2TepqakkJSXh7u6uUQeBNmUQHq2J07x5c44fP05MTAx6enoqM5y0qb8bg7e3N2VlZWzdulXt3xt6vLrqnIqKCtLT0x87T1dUVPDDDz9w584d+vfvLzW6a1Me1YUyUs52asi9qm3btrRu3ZqIiAi111dcXCw9b7q7u6Onp8f+/ftl5enmzZv1rjvREO7u7ujr6xMeHi575r1//z4xMTG0bdtWbRhETeTl5bFo0SIMDAz4+OOPVUKPwaPnHmdnZ06ePCkLk1VRUcG+ffvQ09OjS5cuDTp+bfWbMi/U/Hz37t0NOk59tHlGfZL7AO3yojbatGmDjY0NkZGRasMHK8uYtnXZ3bt3NV6X6Nq1ayxYsAALCwtmzZpVayf7w4cP1aZVSkoKCQkJuLi41Br2Kjs7my1btjB58mQpPzdv3py8vDzpGT83N5e8vDxZJ3pOTg7p6elq75GCIAhC3cRMEUEQ/jaSkpLUdjo4OjqqNAQ0xCuvvMKsWbOYPn06fn5+ODg4UFZWRkZGBvHx8bz11lt07dqV6OhowsPD8fDwkEZXJSYmkp6eLo3sqY2lpSVjxoxh3bp1fPzxx3h5eWFubs7du3c5e/YsX375JQBDhw7lxIkTLFu2jKtXr2Jra8vFixeJi4uje/fuUkPbkzRkyBCSk5P54YcfSEpKomPHjujq6pKZmUlCQgIeHh6MGzcOgLFjx3LhwgWWLFmCj48PTk5OlJSUcPnyZbp06SItAu/s7MyRI0fYtGkTdnZ2KBSKOhcgb9++PQqFgt27d1NWVoaBgQH29vY4ODgwduxYfv31V+bPn4+3tzft27ensrKSe/fucerUKYYPH05gYCAXLlzgp59+wtPTE1tbW/T19bl8+TLJycn4+vpKo/2UjcebNm3Cx8dHWjTUyspK47yho6PDhAkT+Prrr5kxYwb9+/enoqKCgwcP0qpVK3777TeN09/a2pqNGzeSlpaGnZ0dZ8+eJTExER8fH9nipoGBgZw8eZLQ0FAGDBhAYWEhkZGR2NracuPGDdk+J0yYQEhICDNmzCAwMBA9PT2OHj2Kqampyovm0KFDOX78OEuXLiUoKAgrKyvOnj0rvZhVHyXZoUMHhg0bxs6dO0lPT8fDw4OmTZty//59Ll++THl5OaGhoQBSWJHDhw9jYmIidVDZ2toydepUevbsSZs2bTAzMyMjI4MDBw5gY2MjG9WqjBetyQKZ2uY5dQwNDXn99ddZsmQJ7777Ln5+flhaWvLgwQNu3rxJQkICq1atkoVRcHJykkIfGBsb4+npqdGx7OzsCAgI4ODBg8yfP58ePXqQlZXFwYMHsbe3Jy0tTdq2SZMmTJo0iaVLlzJ9+nT8/f1RKBRERUWRnZ3Nf/7zn0Ydie3v709UVBQbNmwgIyMDJycnbty4QVRUFG3btiUgIEDatq7y9Gewt7cnJCREWpA6Ojqae/fu8dprr2mURr169aJp06YcP34cGxubemctKHXv3p3OnTuzb98+cnNzcXFxIS0tjdjYWFq1asXDhw9l2wcGBrJ8+XLmzZuHl5cX2dnZHDp0CAcHB9lsvKeeegozMzPWrFnD77//TrNmzbh06RJXr16ladOmGqdL69atsbW1ZdeuXZSVldGqVSuuX79OfHw81tbWjxWGz9nZmcTERH7++Wc6duyIjo4OPXr00Ko8tWzZkkmTJvHzzz/zwQcf0K9fP5o3b05OTg4pKSncunWL77//HoBu3brRrVs3IiIiePDgAS4uLqSmpnLixInHvpbGYmhoyKxZs/D19aW4uJj9+/ejr6+vcdgfbcog/G8AQUREBHp6evTo0UNldLM29Xdj6Nu3L2fOnGH79u2kpKTQpUsXjIyMyMrKIjk5GRsbG9555x2t91vbvcXZ2Zns7GzeffddfH1961xvoLq0tDSOHTsGPOpEvX37NvHx8eTk5ODh4cHkyZOlbbUpjz/88AO5ubl07tyZli1bUlxcLIVYbEg4HIVCwZtvvsncuXN577338Pf3x8bGhsLCQumcFyxYQOvWrTE1NWXUqFFs2LCB2bNn4+XlRUFBARERETg6OnLz5k2tj18XU1NTRo8ezZo1a5g1axbe3t6UlpYSGRlJZWUlL730UoP3vWLFCu7du0dQUBAXL15U+btybZGJEydKzz2DBg3CyMiI2NhYrl+/zvjx41VmZGqqtvqta9eu6Onp8d133xEUFISBgQFnzpypd12Px6HpM+qT3oc2eVEbOjo6vPbaa4SGhvLRRx8xYMAAWrVqRW5uLgkJCbz22mu0adNG67ps7ty5QP3PkllZWYSGhlJaWoqfn5/K2juGhoZ4eHgAj+qKadOm4eHhQevWrTEyMuLWrVtERUVhYGDApEmTaj3O6tWrcXNzk73Tdu/eHSMjI7744gt8fHyIiYnB2NhYNjhqw4YNREdHs2zZsj/tGUsQBOHvSnSKCILwt5GQkKAyah3A19e3UTpFrKysWLRoEdu3bycxMZEjR45gZGSElZUVgwYNkhqi3dzc+O233zh9+jS5ubno6upiY2PDa6+9Rv/+/es9znPPPUerVq3Yu3cvu3btko5dvdHU1NSUefPmsWnTJk6cOEFBQQEtW7Zk5MiRDB8+/A8JE6Grq8uMGTMIDw8nJiaGxMREdHV1sbCwoHPnznh7e0vbmpmZMX/+fLZu3crp06eJiYnB1NQUJycn2Sjs0aNHk5eXx/79+6WRVHU1UFtaWjJ16lR27drFTz/9xMOHDxk5ciQODg6YmpoSGhrKrl27iIuL48SJE1IjWt++faWXtzZt2tCzZ08uXLhAdHQ0CoUCKysrXnzxRdl6JB07dmT06NEcPnyY77//nsrKSt544w2srKw0zhvK61EoFGzbto1NmzZhYWHB4MGDMTEx4f/+7/80Tv8WLVrw1ltvsXbtWo4cOYKxsTFDhw7lhRdekG3n5ubGlClT2LVrF2vWrMHGxoZJkyaRlpam0inSsWNHZsyYwfr169m2bRumpqb4+fnh5ubGvHnzZNs2a9aMkJAQVq9eTWRkJLq6unTt2pVJkybx9ttvq8RRHjt2LO3atePAgQPs3LmTiooKmjdvTvv27WVraejr6/P222+zefNmVq1aRUVFBb6+vkydOpWnn36a8+fPk5ycTFlZGRYWFgwYMIBhw4bJGq5LSko0XoxV2zxXm549exIaGsqOHTs4fPgwBQUFmJmZYWdnx4QJE9Q2rPv7+7Nq1Sr69u2rVdzpl19+mWbNmnHkyBEuXbqEvb097733HrGxsbJOEXi0fsn06dPZvn0727ZtAx6N1pw0aVKjr2+kq6vLrFmzCAsLIy4ujpiYGMzNzQkKCuKFF16QrTNUV3n6M/Tq1QtLS0t27dolLVZdM9RNXXR1dfH29iY8PFyjel5JoVDwwQcfsH79ek6ePMnp06dp3749M2fO5Mcff6S0tFS2va+vL1lZWRw6dIjVq1fj4ODAO++8Q3R0tKxTxNTUlOnTp7N27Vr27t2Ljo4Obm5uzJkzh5CQEI3Pr0mTJnzyySesXr2aw4cP8/DhQ5ydnfn0009ZtWqV2lG5mho8eDB37twhLi6OQ4cOUVVVxbJlyzA0NNSqPAUGBmJra8uePXvYt28fpaWlNGvWDEdHR6ljXum9995jw4YNxMbGkpCQQPv27ZkxYwbr1q17rGtpLOPHj+f8+fPs3r2bwsJC2rVrx4QJE3BwcNDo+9qUQSVvb28p3aovQl2dpvV3Y1AoFLz11lu4uroSFRVFWFgY8GhEdMeOHetd56s2td1bGjpbLjExkcTERBQKBYaGhrRo0YIuXbrg7e0thb1S0qY8+vj4EBUVRVRUFPn5+ZiYmNC2bVsmT56Mm5tbg87VyclJekaJjY3lwYMHmJqaYmNjw/PPPy8LJTRs2DAMDQ3Zt28f69atw9rampdeeom7d+82eqcIPBpgY25uzt69e9m4cSO6urp06NCB4ODgx5rJqOxk2L9/v9q/K/N6hw4dCAkJISwsjD179lBRUYGdnR3Tpk2rtTxoorb6rVWrVnzyySds3LiRrVu3oq+vT/fu3XnrrbeYMmVKg49XF22eUZ/kPkC7vKgNFxcXQkNDCQsL48iRI5SUlNC8eXM6deok2+eTqMsyMzOlGWgbNmxQ+bulpaXUKaKvr4+HhwfXrl0jLi6OsrIyzM3N8fLyYsSIEbU+u547d47ExESWLl0q+1xZt6xcuZK1a9diZ2fHJ5988pcIBykIgvBPoKj6K6ygKQiCIAjC38KNGzf45JNPGDNmDMOHD//Dj5+amsoHH3zAJ598Ihsp91e0f/9+Vq5cSWhoaKOGsRI0l5mZybRp0xg5ciTBwcGPta81a9YQHh7O8uXLH3ux1srKSl5++WWcnJyYMWPGY+1L+Os7evQoy5cvZ/bs2Q1u+BYEQRAEQRAEofGINUUEQRAEQVBL3bo3ynj9jbmIpjbOnTtHhw4d/vIdIlVVVVLIK9Eh8vdXVlZGdHQ0Xbt21bpDRF0M9cOHD1NYWKgy4lwQBEEQBEEQBEF48kT4LEEQBEEQ1JozZw729va0bduW8vJyEhMTuXjxIr1796Z9+/Z/yjkNGTKEIUOG/CnH1sSDBw84f/4858+fJy0tjWnTpv3ZpyQ8hszMTK5cucLJkyfJz89n2LBhWu9j3bp1ZGZm4urqioGBASkpKcTExGBtbc3AgQMb/6QFQRAEQRAEQRCEOolOEUEQBEEQ1OrRowcnTpwgLi6O8vJyLC0tGTVq1J8SNuvv4vbt23zzzTeYmpoydOjQx4pZLvz5Ll26xPLlyzE3N2fSpEkaL7BeXadOndizZw87d+6kuLgYMzMz+vfvzwsvvKCy6LUgCIIgCIIgCILw5Ik1RQRBEARBEARBEARBEARBEARB+FcQa4oIgiAIgiAIgiAIgiAIgiAIgvCvIDpFBEEQBEEQBEEQBEEQBEEQBEH4VxCdIoIgCIIgCIIgCIIgCIIgCIIg/CuIThFBEARBEARBEARBEARBEARBEP4VRKeIIAiCIAiCIAiCIAiCIAiCIAj/CqJTRBAEQRAEQRAEQRAEQRAEQRCEfwXRKSIIgiAIgiAIgiAIgiAIgiAIwr+C6BQRBEEQBEEQBEEQBEEQBEEQBOFfQXSKCIIgCIIgCIIgCIIgCIIgCILwryA6RQRBEARBEARBEARBEARBEARB+FcQnSKCIAiCIAiCIAiCIAiCIAiCIPwriE4RQRAEQRAEQRAEQRAEQRAEQRD+Ff4fgxc6+qoBYcIAAAAASUVORK5CYII=" }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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" }, "metadata": {}, "output_type": "display_data" } ], "source": [ "calibration_curves(scores, OUTPUT_DIRECTORY)\n", "plt.show()\n", "aggregate_curves(scores, OUTPUT_DIRECTORY)\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "68071958", "metadata": {}, "source": [ "### Primary comparison at 24 original trials\n", "\n", "The aggregate gives each participant equal weight. The unchanged context baselines remain useful references but do not define the primary contrast." ] }, { "cell_type": "code", "execution_count": 26, "id": "9ec6f555", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subject1234Mean
CSP + LDA74.8655.4374.5560.9966.46
Alignment only78.0952.5073.7558.9865.83
Bandpower + LDA59.7635.1760.8751.9551.94
Blended alignment only80.5252.0979.5661.5668.43
Blended alignment + update78.8348.1576.9066.0167.47
Full alignment + update78.5249.0174.6464.7366.73
Frozen tangent decoder77.3950.4780.4355.7466.00
Raw tangent + LR78.2750.4779.9855.7466.11
Supervised update only77.4949.4276.4566.5167.47
\n", "
" ], "text/plain": [ "subject 1 2 3 4 Mean\n", "CSP + LDA 74.86 55.43 74.55 60.99 66.46\n", "Alignment only 78.09 52.50 73.75 58.98 65.83\n", "Bandpower + LDA 59.76 35.17 60.87 51.95 51.94\n", "Blended alignment only 80.52 52.09 79.56 61.56 68.43\n", "Blended alignment + update 78.83 48.15 76.90 66.01 67.47\n", "Full alignment + update 78.52 49.01 74.64 64.73 66.73\n", "Frozen tangent decoder 77.39 50.47 80.43 55.74 66.00\n", "Raw tangent + LR 78.27 50.47 79.98 55.74 66.11\n", "Supervised update only 77.49 49.42 76.45 66.51 67.47" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
subjectestimate_ppci_low_ppci_high_pp
511.3450.4672.202
92-1.264-2.9380.000
1330.4460.0001.415
174-0.500-2.2401.064
21mean0.007-0.6740.607
\n", "
" ], "text/plain": [ " subject estimate_pp ci_low_pp ci_high_pp\n", "5 1 1.345 0.467 2.202\n", "9 2 -1.264 -2.938 0.000\n", "13 3 0.446 0.000 1.415\n", "17 4 -0.500 -2.240 1.064\n", "21 mean 0.007 -0.674 0.607" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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" }, "metadata": {}, "output_type": "display_data" } ], "source": [ "table24=100*scores.query('budget == 24').pivot(index='method', columns='subject', values='balanced_accuracy')\n", "table24['Mean']=table24.mean(axis=1)\n", "table24.index=[DISPLAY_LABELS.get(m,m) for m in table24.index]\n", "display(table24.round(2))\n", "display(intervals.query(\"prespecified_role == 'primary'\")[['subject','estimate_pp','ci_low_pp','ci_high_pp']].round(3))\n", "primary_comparison(intervals, OUTPUT_DIRECTORY)\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "5b427076", "metadata": {}, "source": [ "### Secondary comparisons and the previously observed short-prefix case\n", "\n", "These comparisons were specified before Study 003 execution but were motivated by known Study 002 outcomes. They do not provide independent confirmation. Intervals resample whole test runs and are conditional on these people, fitted models, fixed prefixes and sessions; they do not cover future-day or population variability or correct for the development history." ] }, { "cell_type": "code", "execution_count": 27, "id": "458b5eaf", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
budgetcontrastestimate_ppci_low_ppci_high_pp
48blended_combined_minus_combined5.6643.7737.915
2224blended_combined_minus_combined0.745-0.6942.305
2324blended_combined_minus_frozen1.467-1.7544.204
2424blended_combined_minus_blended_alignment-0.960-3.4591.253
\n", "
" ], "text/plain": [ " budget contrast ... ci_low_pp ci_high_pp\n", "4 8 blended_combined_minus_combined ... 3.773 7.915\n", "22 24 blended_combined_minus_combined ... -0.694 2.305\n", "23 24 blended_combined_minus_frozen ... -1.754 4.204\n", "24 24 blended_combined_minus_blended_alignment ... -3.459 1.253\n", "\n", "[4 rows x 5 columns]" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Participant 3, eight-trial budget (%)
Bandpower + LDA60.87
CSP + LDA74.55
Raw tangent + LR79.98
Frozen tangent decoder80.43
Alignment only49.41
Supervised update only79.53
Full alignment + update52.00
Blended alignment only79.13
Blended alignment + update78.64
\n", "
" ], "text/plain": [ " Participant 3, eight-trial budget (%)\n", "Bandpower + LDA 60.87\n", "CSP + LDA 74.55\n", "Raw tangent + LR 79.98\n", "Frozen tangent decoder 80.43\n", "Alignment only 49.41\n", "Supervised update only 79.53\n", "Full alignment + update 52.00\n", "Blended alignment only 79.13\n", "Blended alignment + update 78.64" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "display(intervals.query(\"subject == 'mean' and prespecified_role == 'secondary'\")[['budget','contrast','estimate_pp','ci_low_pp','ci_high_pp']].round(3))\n", "case=100*scores.query(\"subject == 3 and budget == 8\").set_index('method').balanced_accuracy\n", "case.index=[DISPLAY_LABELS.get(m,m) for m in case.index]\n", "display(case.round(2).to_frame('Participant 3, eight-trial budget (%)'))" ] }, { "cell_type": "markdown", "id": "7bd39235", "metadata": {}, "source": [ "## Takeaways\n", "\n", "**The primary comparison is near zero.** The conservative alignment rule adds no demonstrated benefit over supervised updating alone at the fixed 24-trial budget. The interval does not prove equivalence or rule out effects on other people or days. Leaving out one development participant changes the mean difference from −0.44 to +0.43 points.\n", "\n", "**A known failure was mitigated, not universally prevented.** At eight original trials, participant 3 improved from 52.00% with full alignment plus updating to 78.64% with blended alignment plus updating. Update-only reached 79.53%, and the frozen model reached 80.43%. Across the four people, blended versus full alignment plus updating improved by 5.66 points (conditional 95% interval 3.77 to 7.92), driven largely by participant 3. Participants 1 and 2 worsened by 2.11 and 5.49 points. Participant 2 still scored 43.50% with the blend, below its 50.47% frozen score. These results do not establish broad robustness.\n", "\n", "**Other budgets remain descriptive.** At 48 original trials, update-only averaged 70.17%, versus 69.98% for the blend plus updating. Blended alignment without updating averaged 68.43% at budget 24; the prespecified comparison adding updating was −0.96 points, with an interval spanning zero. Selecting a new winner after inspecting these curves would be another development decision.\n", "\n", "**Research decision.** Keep update-only as the simpler baseline and retain the frozen blend as a comparator. Before further tuning, use a locked protocol for a one-time evaluation on the reserved participants. Those participants can test transfer to new people within this dataset; a later independent dataset and prospective sessions would still be needed for broader claims. This study has not opened the reserved data or executed that next evaluation.\n", "\n", "**Business implication.** We have an auditable evaluation pipeline and a concrete failure mode to investigate. We do not yet have evidence of a superior proprietary decoder or a reduction in real calibration time. The useful product question remains: can an adaptation procedure deliver reliable performance with fewer newly labeled trials?" ] }, { "cell_type": "markdown", "id": "c336b107", "metadata": {}, "source": [ "## Your rerun on another computer\n", "\n", "1. Download and extract the Study 003 bundle into a **new folder**, separate from Study 002. Keep the original results; a rerun overwrites the extracted `results` folder.\n", "2. Open `Neural_Decoder_Experiment_003_Conservative_Alignment.ipynb` in your working Jupyter installation and select **Python 3 (ipykernel)**. Use the pinned `requirements.txt` with Python 3.12 for the closest match. If packages are missing, run `%pip install -r requirements.txt` in a temporary notebook cell, then restart the kernel.\n", "3. Run all cells. The notebook downloads only participants 1–4 if its cache is missing; an initial download may take longer than the computation. Alternatively, set `NEURAL_DATA_DIR` to your existing eight-file cache before running.\n", "4. Copy the final **REFERENCE CHECK** output, including any mismatches or warnings. Send that output back, together with any error traceback. If downloading fails, send the failing URL and error; do not substitute a different dataset.\n", "\n", "This checks computational reproducibility in another environment. It is not validation on new people or recording days. Do not adjust parameters to improve agreement or open participants 5–9. Tiny numerical feature differences across platforms can occur; the check reports scores and exact predicted-label agreement separately.\n", "\n", "## Sources\n", "\n", "- [Dataset and trial timing](https://bnci-horizon-2020.eu/database/data-sets/001-2014/description.pdf)\n", "- [Euclidean alignment: He & Wu](https://arxiv.org/abs/1808.05464)\n", "- [MNE CSP documentation](https://mne.tools/stable/generated/mne.decoding.CSP.html)\n", "- [pyRiemann TangentSpace](https://pyriemann.readthedocs.io/en/latest/generated/pyriemann.tangentspace.TangentSpace.html)\n", "- [scikit-learn LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html)\n", "\n", "The complete bundle includes the protocol, source hashes, pinned packages, standalone code, predictions, access audit, matrices' diagnostic summaries, figures, reference results and independent validation." ] }, { "cell_type": "code", "execution_count": 28, "id": "f36d6fd3", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "REFERENCE CHECK\n", "Score rows: 180 / 180 expected\n", "All scores match within 1e-8: True\n", "Maximum score/count difference: 0\n", "Prediction rows: 40185 / 40185 expected\n", "All predicted labels match exactly: True\n", "PASS: computation reproduces the saved reference.\n" ] } ], "source": [ "# Frozen reference check. These values come from the completed 8 November 2026 run.\n", "import hashlib\n", "import numpy as np\n", "import pandas as pd\n", "REFERENCE_SCORE_COLUMNS = ['subject', 'budget', 'method', 'accuracy', 'balanced_accuracy', 'correct']\n", "REFERENCE_SCORE_ROWS = [[1, 0, 'bandpower_LDA', 0.597457627118644, 0.5975691347011597, 141], [1, 0, 'CSP_LDA', 0.75, 0.7485801367826347, 177], [1, 0, 'raw_tangent_LR', 0.7838983050847458, 0.7826903062741599, 185], [1, 0, 'frozen', 0.7754237288135594, 0.7739183764495987, 183], [1, 0, 'alignment_only', 0.7754237288135594, 0.7739183764495987, 183], [1, 0, 'update_only', 0.7754237288135594, 0.7739183764495987, 183], [1, 0, 'combined', 0.7754237288135594, 0.7739183764495987, 183], [1, 0, 'blended_alignment', 0.7754237288135594, 0.7739183764495987, 183], [1, 0, 'blended_combined', 0.7754237288135594, 0.7739183764495987, 183], [1, 8, 'bandpower_LDA', 0.597457627118644, 0.5975691347011597, 141], [1, 8, 'CSP_LDA', 0.75, 0.7485801367826347, 177], [1, 8, 'raw_tangent_LR', 0.7838983050847458, 0.7826903062741599, 185], [1, 8, 'frozen', 0.7754237288135594, 0.7739183764495987, 183], [1, 8, 'alignment_only', 0.7966101694915254, 0.7978218852215284, 188], [1, 8, 'update_only', 0.7754237288135594, 0.7760407374368125, 183], [1, 8, 'combined', 0.809322033898305, 0.8099539101992268, 191], [1, 8, 'blended_alignment', 0.7966101694915254, 0.7960674992566161, 188], [1, 8, 'blended_combined', 0.788135593220339, 0.7888306571513529, 186], [1, 16, 'bandpower_LDA', 0.597457627118644, 0.5975691347011597, 141], [1, 16, 'CSP_LDA', 0.75, 0.7485801367826347, 177], [1, 16, 'raw_tangent_LR', 0.7838983050847458, 0.7826903062741599, 185], [1, 16, 'frozen', 0.7754237288135594, 0.7739183764495987, 183], [1, 16, 'alignment_only', 0.7838983050847458, 0.7850245316681534, 185], [1, 16, 'update_only', 0.7542372881355932, 0.7538209931608683, 178], [1, 16, 'combined', 0.7711864406779662, 0.7719372584002379, 182], [1, 16, 'blended_alignment', 0.7966101694915254, 0.7965767172167708, 188], [1, 16, 'blended_combined', 0.7754237288135594, 0.7753047873922093, 183], [1, 24, 'bandpower_LDA', 0.597457627118644, 0.5975691347011597, 141], [1, 24, 'CSP_LDA', 0.75, 0.7485801367826347, 177], [1, 24, 'raw_tangent_LR', 0.7838983050847458, 0.7826903062741599, 185], [1, 24, 'frozen', 0.7754237288135594, 0.7739183764495987, 183], [1, 24, 'alignment_only', 0.7796610169491526, 0.7809284864704134, 184], [1, 24, 'update_only', 0.7754237288135594, 0.7748661909009813, 183], [1, 24, 'combined', 0.7838983050847458, 0.7851657746060066, 185], [1, 24, 'blended_alignment', 0.8050847457627118, 0.8051999702646446, 190], [1, 24, 'blended_combined', 0.788135593220339, 0.788314005352364, 186], [1, 48, 'bandpower_LDA', 0.597457627118644, 0.5975691347011597, 141], [1, 48, 'CSP_LDA', 0.75, 0.7485801367826347, 177], [1, 48, 'raw_tangent_LR', 0.7838983050847458, 0.7826903062741599, 185], [1, 48, 'frozen', 0.7754237288135594, 0.7739183764495987, 183], [1, 48, 'alignment_only', 0.7838983050847458, 0.7841473386856973, 185], [1, 48, 'update_only', 0.7966101694915254, 0.7965060957478443, 188], [1, 48, 'combined', 0.7923728813559322, 0.7929898899791854, 187], [1, 48, 'blended_alignment', 0.7966101694915254, 0.7964280404400833, 188], [1, 48, 'blended_combined', 0.809322033898305, 0.8094372584002378, 191], [2, 0, 'bandpower_LDA', 0.350210970464135, 0.3517192674849016, 83], [2, 0, 'CSP_LDA', 0.5527426160337553, 0.554303039158387, 131], [2, 0, 'raw_tangent_LR', 0.5021097046413502, 0.5046512760568868, 119], [2, 0, 'frozen', 0.5021097046413502, 0.5046512760568868, 119], [2, 0, 'alignment_only', 0.5021097046413502, 0.5046512760568868, 119], [2, 0, 'update_only', 0.5021097046413502, 0.5046512760568868, 119], [2, 0, 'combined', 0.5021097046413502, 0.5046512760568868, 119], [2, 0, 'blended_alignment', 0.5021097046413502, 0.5046512760568868, 119], [2, 0, 'blended_combined', 0.5021097046413502, 0.5046512760568868, 119], [2, 8, 'bandpower_LDA', 0.350210970464135, 0.3517192674849016, 83], [2, 8, 'CSP_LDA', 0.5527426160337553, 0.554303039158387, 131], [2, 8, 'raw_tangent_LR', 0.5021097046413502, 0.5046512760568868, 119], [2, 8, 'frozen', 0.5021097046413502, 0.5046512760568868, 119], [2, 8, 'alignment_only', 0.4936708860759494, 0.4939411650107149, 117], [2, 8, 'update_only', 0.430379746835443, 0.4306570231833236, 102], [2, 8, 'combined', 0.4894514767932489, 0.4898475550360414, 116], [2, 8, 'blended_alignment', 0.5147679324894515, 0.5167202415741282, 122], [2, 8, 'blended_combined', 0.4345991561181435, 0.4349673680109098, 103], [2, 16, 'bandpower_LDA', 0.350210970464135, 0.3517192674849016, 83], [2, 16, 'CSP_LDA', 0.5527426160337553, 0.554303039158387, 131], [2, 16, 'raw_tangent_LR', 0.5021097046413502, 0.5046512760568868, 119], [2, 16, 'frozen', 0.5021097046413502, 0.5046512760568868, 119], [2, 16, 'alignment_only', 0.5147679324894515, 0.5161406584843172, 122], [2, 16, 'update_only', 0.4936708860759494, 0.4942333917786869, 117], [2, 16, 'combined', 0.5063291139240507, 0.506947691408533, 120], [2, 16, 'blended_alignment', 0.5189873417721519, 0.5210305864017144, 123], [2, 16, 'blended_combined', 0.5021097046413502, 0.5026373465809468, 119], [2, 24, 'bandpower_LDA', 0.350210970464135, 0.3517192674849016, 83], [2, 24, 'CSP_LDA', 0.5527426160337553, 0.554303039158387, 131], [2, 24, 'raw_tangent_LR', 0.5021097046413502, 0.5046512760568868, 119], [2, 24, 'frozen', 0.5021097046413502, 0.5046512760568868, 119], [2, 24, 'alignment_only', 0.5232067510548524, 0.5250462692382623, 124], [2, 24, 'update_only', 0.4936708860759494, 0.4941603350866939, 117], [2, 24, 'combined', 0.4894514767932489, 0.4901373465809467, 116], [2, 24, 'blended_alignment', 0.5189873417721519, 0.5208869082407949, 123], [2, 24, 'blended_combined', 0.4810126582278481, 0.4815166569257744, 114], [2, 48, 'bandpower_LDA', 0.350210970464135, 0.3517192674849016, 83], [2, 48, 'CSP_LDA', 0.5527426160337553, 0.554303039158387, 131], [2, 48, 'raw_tangent_LR', 0.5021097046413502, 0.5046512760568868, 119], [2, 48, 'frozen', 0.5021097046413502, 0.5046512760568868, 119], [2, 48, 'alignment_only', 0.4978902953586498, 0.4992523865186051, 118], [2, 48, 'update_only', 0.5316455696202531, 0.5326709526592636, 126], [2, 48, 'combined', 0.510548523206751, 0.5111094876290668, 121], [2, 48, 'blended_alignment', 0.5232067510548524, 0.5248319696084162, 124], [2, 48, 'blended_combined', 0.5147679324894515, 0.5156438729787649, 122], [3, 0, 'bandpower_LDA', 0.6123348017621145, 0.6087318401937046, 139], [3, 0, 'CSP_LDA', 0.748898678414097, 0.7455357142857143, 170], [3, 0, 'raw_tangent_LR', 0.801762114537445, 0.7997881355932204, 182], [3, 0, 'frozen', 0.8061674008810573, 0.8042524213075061, 183], [3, 0, 'alignment_only', 0.8061674008810573, 0.8042524213075061, 183], [3, 0, 'update_only', 0.8061674008810573, 0.8042524213075061, 183], [3, 0, 'combined', 0.8061674008810573, 0.8042524213075061, 183], [3, 0, 'blended_alignment', 0.8061674008810573, 0.8042524213075061, 183], [3, 0, 'blended_combined', 0.8061674008810573, 0.8042524213075061, 183], [3, 8, 'bandpower_LDA', 0.6123348017621145, 0.6087318401937046, 139], [3, 8, 'CSP_LDA', 0.748898678414097, 0.7455357142857143, 170], [3, 8, 'raw_tangent_LR', 0.801762114537445, 0.7997881355932204, 182], [3, 8, 'frozen', 0.8061674008810573, 0.8042524213075061, 183], [3, 8, 'alignment_only', 0.4933920704845815, 0.4940980629539951, 112], [3, 8, 'update_only', 0.7973568281938326, 0.7953238498789346, 181], [3, 8, 'combined', 0.5198237885462555, 0.5199757869249395, 118], [3, 8, 'blended_alignment', 0.7929515418502202, 0.791313559322034, 180], [3, 8, 'blended_combined', 0.788546255506608, 0.7863952784503633, 179], [3, 16, 'bandpower_LDA', 0.6123348017621145, 0.6087318401937046, 139], [3, 16, 'CSP_LDA', 0.748898678414097, 0.7455357142857143, 170], [3, 16, 'raw_tangent_LR', 0.801762114537445, 0.7997881355932204, 182], [3, 16, 'frozen', 0.8061674008810573, 0.8042524213075061, 183], [3, 16, 'alignment_only', 0.748898678414097, 0.7466707021791767, 170], [3, 16, 'update_only', 0.788546255506608, 0.7863952784503632, 179], [3, 16, 'combined', 0.762114537444934, 0.760063559322034, 173], [3, 16, 'blended_alignment', 0.7973568281938326, 0.7955508474576272, 181], [3, 16, 'blended_combined', 0.7797356828193832, 0.7774667070217918, 177], [3, 24, 'bandpower_LDA', 0.6123348017621145, 0.6087318401937046, 139], [3, 24, 'CSP_LDA', 0.748898678414097, 0.7455357142857143, 170], [3, 24, 'raw_tangent_LR', 0.801762114537445, 0.7997881355932204, 182], [3, 24, 'frozen', 0.8061674008810573, 0.8042524213075061, 183], [3, 24, 'alignment_only', 0.7400881057268722, 0.737515133171913, 168], [3, 24, 'update_only', 0.7665198237885462, 0.7645278450363195, 174], [3, 24, 'combined', 0.748898678414097, 0.7464437046004841, 170], [3, 24, 'blended_alignment', 0.7973568281938326, 0.7955508474576272, 181], [3, 24, 'blended_combined', 0.7709251101321586, 0.7689921307506054, 175], [3, 48, 'bandpower_LDA', 0.6123348017621145, 0.6087318401937046, 139], [3, 48, 'CSP_LDA', 0.748898678414097, 0.7455357142857143, 170], [3, 48, 'raw_tangent_LR', 0.801762114537445, 0.7997881355932204, 182], [3, 48, 'frozen', 0.8061674008810573, 0.8042524213075061, 183], [3, 48, 'alignment_only', 0.788546255506608, 0.7866222760290557, 179], [3, 48, 'update_only', 0.8061674008810573, 0.8044794188861986, 183], [3, 48, 'combined', 0.801762114537445, 0.8002421307506052, 182], [3, 48, 'blended_alignment', 0.7973568281938326, 0.7955508474576272, 181], [3, 48, 'blended_combined', 0.801762114537445, 0.8002421307506054, 182], [4, 0, 'bandpower_LDA', 0.5181347150259067, 0.5194919669995658, 100], [4, 0, 'CSP_LDA', 0.6062176165803109, 0.6099435518888406, 117], [4, 0, 'raw_tangent_LR', 0.5595854922279793, 0.5573599652627008, 108], [4, 0, 'frozen', 0.5595854922279793, 0.5573599652627008, 108], [4, 0, 'alignment_only', 0.5595854922279793, 0.5573599652627008, 108], [4, 0, 'update_only', 0.5595854922279793, 0.5573599652627008, 108], [4, 0, 'combined', 0.5595854922279793, 0.5573599652627008, 108], [4, 0, 'blended_alignment', 0.5595854922279793, 0.5573599652627008, 108], [4, 0, 'blended_combined', 0.5595854922279793, 0.5573599652627008, 108], [4, 8, 'bandpower_LDA', 0.5181347150259067, 0.5194919669995658, 100], [4, 8, 'CSP_LDA', 0.6062176165803109, 0.6099435518888406, 117], [4, 8, 'raw_tangent_LR', 0.5595854922279793, 0.5573599652627008, 108], [4, 8, 'frozen', 0.5595854922279793, 0.5573599652627008, 108], [4, 8, 'alignment_only', 0.5647668393782384, 0.5679765523230569, 109], [4, 8, 'update_only', 0.6424870466321243, 0.6398610508033, 124], [4, 8, 'combined', 0.5958549222797928, 0.5989339991315675, 115], [4, 8, 'blended_alignment', 0.5854922279792746, 0.5837386018237083, 113], [4, 8, 'blended_combined', 0.6373056994818653, 0.6350781589231437, 123], [4, 16, 'bandpower_LDA', 0.5181347150259067, 0.5194919669995658, 100], [4, 16, 'CSP_LDA', 0.6062176165803109, 0.6099435518888406, 117], [4, 16, 'raw_tangent_LR', 0.5595854922279793, 0.5573599652627008, 108], [4, 16, 'frozen', 0.5595854922279793, 0.5573599652627008, 108], [4, 16, 'alignment_only', 0.5803108808290155, 0.5839339991315675, 112], [4, 16, 'update_only', 0.6683937823834197, 0.667822405557968, 129], [4, 16, 'combined', 0.6373056994818653, 0.6412700825010855, 123], [4, 16, 'blended_alignment', 0.5958549222797928, 0.5942618323925315, 115], [4, 16, 'blended_combined', 0.694300518134715, 0.6940990013026487, 134], [4, 24, 'bandpower_LDA', 0.5181347150259067, 0.5194919669995658, 100], [4, 24, 'CSP_LDA', 0.6062176165803109, 0.6099435518888406, 117], [4, 24, 'raw_tangent_LR', 0.5595854922279793, 0.5573599652627008, 108], [4, 24, 'frozen', 0.5595854922279793, 0.5573599652627008, 108], [4, 24, 'alignment_only', 0.5854922279792746, 0.5897894051237516, 113], [4, 24, 'update_only', 0.6632124352331606, 0.6650564481111594, 128], [4, 24, 'combined', 0.6424870466321243, 0.6473295701259226, 124], [4, 24, 'blended_alignment', 0.616580310880829, 0.6156404689535389, 119], [4, 24, 'blended_combined', 0.6580310880829016, 0.6600564481111594, 127], [4, 48, 'bandpower_LDA', 0.5181347150259067, 0.5194919669995658, 100], [4, 48, 'CSP_LDA', 0.6062176165803109, 0.6099435518888406, 117], [4, 48, 'raw_tangent_LR', 0.5595854922279793, 0.5573599652627008, 108], [4, 48, 'frozen', 0.5595854922279793, 0.5573599652627008, 108], [4, 48, 'alignment_only', 0.5440414507772021, 0.5501606600086844, 105], [4, 48, 'update_only', 0.6735751295336787, 0.6732826747720364, 130], [4, 48, 'combined', 0.6217616580310881, 0.626079027355623, 120], [4, 48, 'blended_alignment', 0.616580310880829, 0.6173643074250976, 119], [4, 48, 'blended_combined', 0.6735751295336787, 0.6738189318280503, 130]]\n", "REFERENCE_PREDICTION_HASH = 'c16c0837b093b58978dd5bddf7b22ab4d76d6362109f6b9eb086cca6933a9b93'\n", "REFERENCE_PREDICTION_ROWS = 40185\n", "_ref = pd.DataFrame(REFERENCE_SCORE_ROWS, columns=REFERENCE_SCORE_COLUMNS)\n", "_actual = pd.read_csv(OUTPUT_DIRECTORY / 'scores.csv')\n", "_keys = ['subject', 'budget', 'method']\n", "_joined = _actual.merge(_ref, on=_keys, how='outer', suffixes=('_actual', '_reference'), indicator=True, validate='one_to_one')\n", "_all_keys_match = bool((_joined['_merge'] == 'both').all())\n", "_metrics = ['accuracy', 'balanced_accuracy', 'correct']\n", "_max_score_error = max(float((_joined[m + '_actual'] - _joined[m + '_reference']).abs().max()) for m in _metrics)\n", "_scores_match = _all_keys_match and all(np.allclose(_joined[m + '_actual'], _joined[m + '_reference'], rtol=0, atol=1e-8) for m in _metrics)\n", "_pred = pd.read_csv(OUTPUT_DIRECTORY / 'predictions.csv')\n", "_pred_columns = ['subject', 'budget', 'method', 'trial_id', 'label', 'prediction']\n", "_pred_keys = ['subject', 'budget', 'method', 'trial_id']\n", "_digest = hashlib.sha256(_pred[_pred_columns].sort_values(_pred_keys).to_csv(index=False, lineterminator='\\n').encode()).hexdigest()\n", "_predictions_match = len(_pred) == REFERENCE_PREDICTION_ROWS and _digest == REFERENCE_PREDICTION_HASH\n", "print('REFERENCE CHECK')\n", "print(f'Score rows: {len(_actual)} / {len(_ref)} expected')\n", "print(f'All scores match within 1e-8: {_scores_match}')\n", "print(f'Maximum score/count difference: {_max_score_error:.3g}')\n", "print(f'Prediction rows: {len(_pred)} / {REFERENCE_PREDICTION_ROWS} expected')\n", "print(f'All predicted labels match exactly: {_predictions_match}')\n", "print('PASS: computation reproduces the saved reference.' if _scores_match and _predictions_match else 'DIFFERENCE: save this output and warnings for review; do not tune parameters.')\n" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "name": "python", "version": "3.12.14" }, "study_id": "Daybreak Neural 003" }, "nbformat": 4, "nbformat_minor": 5 }