{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Playlist Neural Network\n", "\n", "Given a list of playlists, can unknown tracks be correctly classified?" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "# playlist_names = [\"RAP\", \"EDM\", \"ROCK\", \"METAL\", \"JAZZ\", \"POP\"] # super-genres\n", "# playlist_names = [\"ALL RAP\", \"EDM\", \"ROCK\", \"METAL\", \"JAZZ\", \"POP\"] # super-genres\n", "# playlist_names = [\"RAP\", \"EDM\", \"ROCK\", \"METAL\", \"JAZZ\"] # super-genres without POP\n", "# playlist_names = [\"ALL RAP\", \"EDM\", \"ROCK\", \"METAL\", \"JAZZ\"] # super-genres without POP\n", "playlist_names = [\"ALL RAP\", \"DNB\", \"4/4\", \"cRock\", \"METAL\", \"cJazz\"] # super-genres with decomposed EDM\n", "# playlist_names = [\"DNB\", \"HOUSE\", \"TECHNO\", \"GARAGE\", \"DUBSTEP\", \"BASS\"] # EDM playlists\n", "# playlist_names = [\"20s rap\", \"10s rap\", \"00s rap\", \"90s rap\", \"80s rap\"] # rap decades\n", "# playlist_names = [\"UK RAP\", \"US RAP\"] # UK/US split\n", "# playlist_names = [\"uk rap\", \"grime\", \"drill\", \"afro bash\"] # british rap playlists\n", "# playlist_names = [\"20s rap\", \"10s rap\", \"00s rap\", \"90s rap\", \"80s rap\", \"trap\", \"gangsta rap\", \"industrial rap\", \"weird rap\", \"jazz rap\", \"boom bap\", \"trap metal\"] # american rap playlists\n", "# playlist_names = [\"rock\", \"indie\", \"punk\", \"pop rock\", \"bluesy rock\", \"hard rock\", \"chilled rock\", \"emo\", \"pop punk\", \"stoner rock/metal\", \"post-hardcore\", \"melodic hardcore\", \"art rock\", \"post-rock\", \"classic pop punk\", \"90s rock & grunge\", \"90s indie & britpop\", \"psych\"] # rock playlists\n", "# playlist_names = [\"metal\", \"metalcore\", \"mathcore\", \"hardcore\", \"black metal\", \"death metal\", \"doom metal\", \"sludge metal\", \"classic metal\", \"industrial\", \"nu metal\", \"calm metal\", \"thrash metal\"] # metal playlists\n", "\n", "# headers = float_headers + [\"duration_ms\", \"mode\", \"loudness\", \"tempo\"]\n", "headers = float_headers + [\"mode\", \"loudness\", \"tempo\"]\n", "# headers = float_headers\n", "\n", "BALANCED_WEIGHTS = True" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Pull and process playlist information.\n", "\n", "1. Get live playlist track information from spotify\n", "2. Filter listening history for these tracks\n", "\n", "Filter out tracks without features and drop duplicates before taking only the descriptor parameters" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [], "source": [ "playlists = [get_playlist(i, spotnet) for i in playlist_names] # 1)\n", "\n", "# filter playlists by join with playlist track/artist names\n", "filtered_playlists = [pd.merge(track_frame(i.tracks), scrobbles, on=['track', 'artist']) for i in playlists] # 2)\n", "\n", "filtered_playlists = [i[pd.notnull(i[\"uri\"])] for i in filtered_playlists]\n", "# distinct on uri\n", "filtered_playlists = [i.drop_duplicates(['uri']) for i in filtered_playlists]\n", "# select only descriptor float columns\n", "filtered_playlists = [i[headers] for i in filtered_playlists]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Construct the dataset with associated labels before splitting into a train and test set." ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [], "source": [ "dataset = pd.concat(filtered_playlists)\n", "labels = [np.full(len(plst), idx) for idx, plst in enumerate(filtered_playlists)]\n", "labels = np.concatenate(labels)\n", "\n", "# stratify: maintains class proportions in test and train set\n", "data_train, data_test, labels_train, labels_test = train_test_split(dataset, labels, \n", " test_size=0.1, \n", "# random_state=70, \n", " stratify=labels\n", " )\n", "\n", "class_weights = class_weight.compute_class_weight('balanced',\n", " classes=np.unique(labels_train),\n", " y=labels_train)\n", "class_weights = {i: j for i, j in zip(range(len(filtered_playlists)), class_weights)}\n", "\n", "labels_train = tf.one_hot(labels_train, len(filtered_playlists))\n", "labels_test = tf.one_hot(labels_test, len(filtered_playlists))" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [], "source": [ "def tensorboard_callback(path='tensorboard-logs', prefix=''):\n", " return tf.keras.callbacks.TensorBoard(\n", " log_dir=os.path.normpath(os.path.join(path, prefix + datetime.now().strftime(\"%Y%m%d-%H%M%S\"))), histogram_freq=1\n", " )" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [], "source": [ "def get_model(hidden_nodes=128,\n", " layers=2,\n", " classes=len(filtered_playlists),\n", " activation=lambda: 'sigmoid', \n", " weight_init=lambda: 'glorot_uniform'):\n", " l = [tf.keras.layers.InputLayer(input_shape=data_train.to_numpy()[0].shape, name='Input')]\n", " \n", " for i in range(layers):\n", " l.append(\n", " tf.keras.layers.Dense(hidden_nodes, \n", " activation=activation(), \n", " kernel_initializer=weight_init(), \n", " name=f'Hidden{i+1}')\n", " )\n", " \n", " l.append(tf.keras.layers.Dense(classes, \n", " activation='softmax', \n", " kernel_initializer=weight_init(), \n", " name='Output'))\n", " \n", " model = tf.keras.models.Sequential(l)\n", " return model" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Single Model" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model: \"sequential_10\"\n", "_________________________________________________________________\n", "Layer (type) Output Shape Param # \n", "=================================================================\n", "Hidden1 (Dense) (None, 64) 704 \n", "_________________________________________________________________\n", "Output (Dense) (None, 6) 390 \n", "=================================================================\n", "Total params: 1,094\n", "Trainable params: 1,094\n", "Non-trainable params: 0\n", "_________________________________________________________________\n" ] } ], "source": [ "model = get_model(hidden_nodes=64, layers=1)\n", "\n", "model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.01), \n", "# optimizer=tf.keras.optimizers.SGD(learning_rate=0.01, momentum=0.9),\n", " loss='categorical_crossentropy', \n", " metrics=['accuracy'])\n", "model.summary()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Train" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [], "source": [ "if BALANCED_WEIGHTS:\n", " cw = class_weights\n", "else:\n", " cw = None\n", "history = model.fit(data_train.to_numpy(), labels_train, \n", " callbacks=[tensorboard_callback()], \n", " validation_split=0.11,\n", " verbose=0,\n", " class_weight=cw,\n", " epochs=50)" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAoQAAAG/CAYAAADB4sa8AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAABJ0AAASdAHeZh94AAB72klEQVR4nO3dd3hUVf7H8fdJgFBCE+mCINJEmigqAoINC+qu4lqw66o/166ra13b6trWLmvvKIptrVhBFAQEAUGKSJUqHUIIKef3x5mbO0kmydTMJPN5Pc88uXPrmZxk5jvnnO+5xlqLiIiIiKSvjGQXQERERESSSwGhiIiISJpTQCgiIiKS5hQQioiIiKQ5BYQiIiIiaU4BoYiIiEiaU0AoIiIikuYUEIqIiIikOQWEIiIiImlOAaGIiIhImlNAKCIiIpLmFBCKiIiIpLmUDgiNMdnGmDuMMZ8ZYzYaY6wx5twIjm9ijHnGGPOHMSbHGPONMWa/BBZZREREpNpJ6YAQ2B24DegOzIrkQGNMBvAxcAbwBHA90AIYb4zpHOdyioiIiFRbtZJdgEqsBlpba9cYY/YHpkVw7AhgAHCKtXYsgDHmLWAhcAcuUBQRERFJeyndQmitzbPWrony8BHAWuDdoPP9AbwFnGiMyYpDEUVERESqvVRvIYxFX2CGtbao1PqpwEVAF+DnUAcaY1oAzUutzg4cMwfYFd+iioiIiMRVHaAdMMFau6WynWtyQNga+DbE+tWBn20oJyAELgX+mYhCiYiIiFShE4H/VbZTTQ4I6wF5IdbvDNpenqeAt0ut6waMHT16NHvttVccildWbm4uP//8Mz179qRevYqKJ8mg+kldqpvUpvpJXaqb1BZL/SxevJgzzjgDYEU4+9fkgDAXCDVOsG7Q9pCsteuAdcHrjDEA9OrVix49esSpiCVt3bqVLVu2sN9++9GoUaOEXEOip/pJXaqb1Kb6SV2qm9QWS/1kZ2d7i2ENc0vppJIYrcZ1G5fmrVtVhWURERERSVk1OSCcCewXmI8w2IHADtz0MyIiIiJpr0YEhMaY1saYbsaY2kGrxwItgZOC9tsdOAX40FobanyhiIiISNpJ+TGExpjLgCa4rGCA440xewSWHw+kUt8LnAN0BJYGto0FfgBeNMbsA6zHZQ9nogxiERERkWIpHxAC1wF7Bj0/Cb/V7zUg5Nw61tpCY8yxwAPAFbis4mnAudbaBYkrroiIiEj1kvIBobW2Qxj7nAucG2L9JuDCwENERCQlFRUVsXbtWvLy8igqKn0/hcTJz89nt912Y/Xq1axfv77KrivhKV0/GRkZZGVl0bJlSzIy4jvqL+UDQhERkZqsqKiI5cuXk5ubS2ZmJpmZmcVTnSVarVq1aN68ObVqKRxIRcH1Y61l165d5ObmkpeXR/v27eMaFOovQEREJInWrl1Lbm4uu+22Gy1atKiyYBCgsLCQbdu20bBhQzIzM6vsuhKe0vVjrWXdunVs3LiRtWvX0rp1qNn1olMjsoxFRESqq7y8PDIzM6s8GJTqxxhDixYtyMzMJC8vvpOlKCAUERFJoqKioirtJpbqzRhDZmZm3MeaKiAUERFJMgWDEolE/L0oIBQRERFJcwoIRURERNKcAkIRERGpts4991w6dOiQ7GJUewoIRUREJO6MMWE9xo8fn+yiCpqHUERERBLg1VdfLfH8lVde4Ysvviizvnv37jFd59lnn63Su7vUVAoIRUREJO7OPPPMEs9/+OEHvvjiizLrS9uxYwf169cP+zq1a9eOqnxSkrqMRUREJCmGDBnCvvvuy/Tp0xk8eDD169fnpptuAuCDDz7guOOOo02bNmRlZdGpUyfuuusuCgsLS5yj9BjCpUuXYozhwQcf5JlnnqFTp05kZWVxwAEHMG3atKp8edWKWghFREQkaTZs2MAxxxzDaaedxplnnknLli0BeOmll8jOzuaaa64hOzubr7/+mttuu42tW7fywAMPVHre0aNHs23bNi6++GKMMdx///2cdNJJLF68WK2KISggFBERSUFXXQUzZyb2GtZmUFiYTWZmBpXNddynDzzySPzLsGbNGv773/9y8cUXl1g/evRo6tWrV/z8kksu4ZJLLuGpp57i7rvvJisrq8LzLl++nF9//ZWmTZsC0LVrV0488UTGjRvH8OHD4/9CqjkFhCIiIilo5kyYMCHRVzEkOxTIysrivPPOK7M+OBjctm0beXl5DBo0iKeffpr58+fTu3fvCs976qmnFgeDAIMGDQJg8eLFcSp5zaKAUEREJAX16ZP4a1hrKSwsDOteyokqT9u2balTp06Z9XPnzuWWW27h66+/ZuvWrSW2bdmypdLztm/fvsRzLzjctGlTDKWtuRQQioiIpKBEdM+WVlhYxLZt22nYsCGZmZmJv2AIwS2Bns2bN3PooYfSqFEj7rzzTjp16kTdunWZMWMGN9xwQ1jTzJT3eqy1MZe5JlJAKCIiIill/PjxbNiwgXfffZfBgwcXr1+yZEkSS1WzadoZERERSSle615wa96uXbt46qmnklWkGk8thCIiIpJSBgwYQNOmTTnnnHO44oorMMbw6quvqrs3gdRCKCIiIimlWbNmfPTRR7Ru3ZpbbrmFBx98kCOPPJL7778/2UWrsdRCKCIiIgn3xBNP8MQTT5RYN378+HL3HzBgAJMnTy6zvnQr4UsvvVTieYcOHcptSVQLY/nUQigiIiKS5hQQioiIiKQ5BYQiIiIiaU4BoYiIiEiaU0AoIiIikuYUEIqIiIikOQWEIiIiImlOAaGIiIhImlNAKCIiIpLmFBCKiIiIpDkFhCIiIiJpTgGhiIiIpLylS5dijClx7+Lbb78dY0xYxxtjuP322+NapiFDhjBkyJC4njNZFBCKiIhI3J1wwgnUr1+fbdu2lbvPyJEjqVOnDhs2bKjCkkXml19+4fbbb2fp0qXJLkpCKSAUERGRuBs5ciS5ubm89957Ibfv2LGDDz74gKOPPppmzZpFdY1bbrmF3NzcWIpZqV9++YU77rgjZED4+eef8/nnnyf0+lVFAaGIiIjE3QknnEDDhg0ZPXp0yO0ffPABOTk5jBw5Mupr1KpVi7p160Z9fKzq1KlDnTp1knb9eFJAKCIiInFXr149TjrpJL766ivWrVtXZvvo0aNp2LAhAwcO5LrrrqNnz55kZ2fTqFEjjjnmGGbNmlXpNUKNIczLy+Pqq6+mefPmNGzYkBNOOIHff/+9zLHLli3j0ksvpWvXrtSrV49mzZpxyimnlGgJfOmllzjllFMAGDp0KMYYjDGMHz8eCD2GcN26dVxwwQW0bNmSunXr0rt3b15++eUS+3jjIR988EGeeeYZOnXqRFZWFgcccADTpk2r9HUnQq2kXFVERERqvJEjR/Lyyy/z1ltvcdlllxWv37hxI+PGjeP0009n9erVvP/++5xyyil07NiRtWvX8vTTT3PooYfyyy+/0KZNm4iueeGFF/Laa69xxhlnMGDAAL7++muOO+64MvtNmzaNSZMmcdppp7HHHnuwdOlSRo0axZAhQ/jll1+oX78+gwcP5oorruCxxx7jpptuonv37gDFP0vLzc1lyJAhLFq0iMsuu4yOHTvy9ttvc+6557J582auvPLKEvuPHj2abdu2cfHFF2OM4f777+ekk05i8eLF1K5dO6LXHSsFhCIiIpIQhx12GK1bt2b06NElAsK3336b/Px8Ro4cSc+ePVm4cCEZGX6n5VlnnUW3bt14/vnnufXWW8O+3qxZs3jttde49NJLefLJJwH429/+xsiRI5k9e3aJfY877jhGjBhRYt3xxx/PwQcfzDvvvMNZZ53FXnvtxaBBg3jsscc48sgjK80ofuaZZ5g3bx6vvfZacVf4JZdcwqGHHsott9zC+eefT8OGDYv3X758Ob/++itNmzYFoGvXrpx44omMGzeO4cOHh/2640EBoYiISCq66iqYOTOhl8iwluzCQjIyM6Gy6Vv69IFHHono/JmZmZx22mk8/PDDLF26lA4dOgCuZaxly5YcfvjhZGZmFu9fWFjI5s2byc7OpmvXrsyYMSOi633yyScAXHHFFSXWX3XVVWXGMtarV694OT8/n61bt7L33nvTpEkTZsyYwVlnnRXRtb3rt2rVitNPP714Xe3atbniiis4/fTTmTBhQolA79RTTy0OBgEGDRoEwOLFiyO+dqwUEIqIiKSimTNhwoSEXsKQ+EBg5MiRPPzww4wePZqbbrqJ33//nYkTJ3LFFVeQmZlJUVERjz76KE899RRLliyhsLCw+NhIs4+XLVtGRkYGnTp1KrG+a9euZfbNzc3l3nvv5cUXX2TlypVYa4u3bdmyJcJX6V+/c+fOJVo7we9iXrZsWYn17du3L/HcCw43bdoU1fVjoYBQREQkFfXpk/BLWGspLCwkMzOz8gmeoyxPv3796NatG2+88QY33XQTb7zxBtba4i7Ve+65h1tvvZXzzz+fu+66i912242MjAyuuuoqioqKorpmOC6//HJefPFFrrrqKg4++GAaN26MMYbTTjstodcNFtw6Giw4OK0qCghFRERSUYTds9EoKixk+7ZtNGzYsNzgJB5GjhzJrbfeyuzZsxk9ejSdO3fmgAMOAGDs2LEMHTqU559/vsQxmzdvZvfdd4/oOnvuuSdFRUX89ttvJVoFFyxYUGbfsWPHcs455/DQQw8Vr9u5cyebN28usV+4d0Lxrj979myKiopKtBLOnz+/eHuq0rQzIiIiklBea+Btt93GzJkzS8w9mJmZWaZF7O2332blypURX+eYY44B4LHHHiux/pEQwXWo6z7++OMluqwBGjRoAFAmUAzl2GOPZc2aNYwZM6Z4XUFBAY8//jjZ2dkceuih4byMpFALoYiIiCRUx44dGTBgAB988AFAiYBw+PDh3HnnnZx33nkMGDCAn3/+mddff5299tor4uv06dOH008/naeeeootW7YwYMAAvvrqKxYtWlRm3+HDh/Pqq6/SuHFj9tlnHyZPnsyXX35ZZtxinz59yMzM5L777mPLli1kZWVx2GGH0aJFizLnvOiii3j66ac599xzmT59Oh06dGDs2LF8//33PPLIIyUyjFONAkIRERFJuJEjRzJp0iT69+/P3nvvXbz+pptuIicnh9GjRzNmzBj2228/Pv74Y/7xj39EdZ0XXniB5s2b8/rrr/P+++9z2GGH8fHHH9OuXbsS+z366KNkZmby+uuvs3PnTg455BC+/PJLhg0bVmK/Vq1a8d///pd7772XCy64gMLCQr755puQAWG9evUYP348//jHP3j55ZfZunUrXbt25cUXX+Tcc8+N6vVUFZOMgYvVkTGmBzBnzpw59OjRIyHX2Lp1K9988w1Dhw6lUaNGCbmGRE/1k7pUN6lN9VMxb4qRaFrEYlVYWMi2KhhDKNEpr37C+ZuZO3cu++67L8C+1tq5lV1LYwhFRERE0pwCQhEREZE0p4BQREREJM0pIBQRERFJcwoIRURERNKcAkIREZEk04wfEolE/L0oIBQREUmijIwMCgsLFRRKWLz7TwffGi8eFBCKiIgkUVZWFoWFhaxbt05BoVTIWsu6desoLCwkKysrrufWnUpERESSqGXLluTl5bFx40a2bNlCZmYmxpgquba1lvz8fDZs2FBl15TwBdcPuImqCwsLqVevHi1btozrtdRCKCIikkQZGRm0b9+eJk2aUKdOnSoNzAoKCvjjjz8oKCiosmtK+ILrxxhDnTp1aNKkCe3bt497l3FKtxAaY7KAO4GzgKbAbOAWa+0XYRx7BHAz0BP3OhcCj1trX01ciUVERCKXkZFB69atq/y6W7duZf78+fTu3Vu3FUxBVVk/qd5C+BJwDfA6cCVQCHxijBlY0UHGmBOAz4E6wO24wDAXeMUYc3UCyysiIiJS7aRsQGiM6Q+cBtxorf27tfYZ4DBgGXB/JYdfBqwGDrPWPmGtfRI4HPgNODdxpRYREUmQyZPhrLNg5sxkl0RqoFTuMh6BaxF8xlthrd1pjHkeuMcY085au6KcYxsBm6y1eUHHFhhj1ie0xCIiIolyzTXwww/wxx/w2WfJLo3UMCnbQgj0BRZaa7eWWj818LNPBceOB3oYY+4yxuxtjOlkjLkV2J/KWxdFRERSS1ERzJrllqdMAU1PI3GWyi2ErXHdvqV569pUcOxdQEfc2MFbAut2ACdbaz+o7MLGmBZA81KrOwFs376drVtLx6jxkZOTU+KnpBbVT+pS3aQ21U/szOLFNMzNdU82b2bbrFnYvfaK+byqm9QWS/1s3749ov1TOSCsB+SFWL8zaHt58nBZxWOBd4FM4CLgNWPMkdbaHyq59qXAP0NtmDp1KmvWrKnk8NhMnTq18p0kaVQ/qUt1k9pUP9FrNXUqBwY9n//666waWGF+ZURUN6ktmvpZvnx5RPunckCYC4Sahrtu0PbyPAEcBOxnrS0CMMa8BcwFHoUS/1ehPAW8XWpdJ+CD/v37071790oOj05OTg5Tp06lf//+NGjQICHXkOipflKX6ia1qX5iV2fGjBLPe+Xn03Xo0JjPq7pJbbHUz7x58yLaP5UDwtVA2xDrvYmaVoU6yBhTB7gAuN8LBgGstfnGmE+By4wxday1u8q7sLV2HbCu1HkByM7OTvhcQA0aNNB8UClM9ZO6VDepTfUTg99+K/E0a84csuL4u1TdpLZo6ic7Ozui/VM5qWQm0MUYU/o3cGDQ9lCa4QLdzBDbauNec6htIiIiqWnu3JLPp09XYonEVSoHhGPxx/4BxXcuOQ+Y4k05Y4xpb4zpFnTcOmAz8OdAa6F3bDZwPDDfWltRd7OIiEjqKCyE+fPdcsOG7ufmzbB4cdKKJDVPygaE1topuHF89xpj7jfGXAR8DXQArg/a9RVgXtBxhcCDQBfgB2PMVcaYa3HT1ewB3F01r0BERCQOFi+GnYF8yr/8xV8/fXpyyiM1UsoGhAFnA4/g7mX8GK7Ld7i19tuKDrLW/gsYCeTjsoXvArYCI6y1ryeywCIiInEV3F18+umQGRj1pIBQ4iiVk0qw1u4E/h54lLfPkHLWjwZGJ6ZkIilo2zZ49lmoWxcuuQQyUv37noRUWAiPPgrt28OIEckujaSC4ICwXz/o0QNmz1ZAKHGV0gGhiITBWhgzBq69Flat8tf97W/JLZdE55FH4LrroFYtGDAA2lQ0B381UFBA5pdfUlsTH0fPCwjbtoUmTVxQ6AWE1kJgFgyRWKgJQaQ6mzMHDjvMdSOtCpqJ6R//gAgnJZUUsGMH3B+4u2ZBAfzyS3LLEw8PPkiDk0/mkNtuc7dfk8h5AWGPHu5nv37upxJLJI4UEIpUR1u2wNVXQ58+MH68W9e8uWtZAti+3XUba1qK6uXpp2Fd0BSopeaeq5b+9z8AGi9dSq3PP09yYaqhggI/w7h0QAjqNpa4UUAoUp0UFcErr0DXrq5rsbDQjRW84gpYuBAeeADOOcft++mnMLqKhtEWFsKf/wyDB0OC7vVd4+Xm+q2DnuoeEO7aBUF32KjzxBNJLEw19dtv7vcIfkDYu7cSSyTuFBCKVBczZ8KgQS7gW7vWrRs0CH76ySUhNGni1v3nP9CihVu+8kr444/El+2HH+D992HiRDeeUSL37LPg3SfdGxO2aFHyyhMPs2dDnn9L+loTJ7q/YwlfcEKJFxDWq+cvKyCUOFFAKJLqduxwCSL9+sGkSW5d69bw+uswYQL06lVy/912g8cfd8sbNsBVVyW+jMFj3SZOTPz1apqdO+G++9zy3nvDkUe65ereQvjDD2XXPfJIlRcjrn7/HV54wWX1V4XggHCfffxlr9tYdyypllJxOK0CQpFUd+WV8NRT7h2kVi2XTTx/PpxxRvnZhaecAiee6JZHj4aPP05sGYNvov5thdOESijPP+8nBd18sxsSAC4grM4f9lOmAFC0226sOuggt+6NN/yW0OqmoACGDYMLLnA/vcmiE8kLCNu1g+B72SqxJCoFBe7t8LTTYL/94Pzz4cUX4ddfq+5fbdw46N/f7+hJFQoIRVLZggWuNQLcB8CsWfDggyU/GEIxBp580t/vkksSO7YvuIVw2TJYsSJx16pp8vLg3/92yx07wsiR0KmTe56TUzLJpLoJBISF/fqx6E9/cut27XJfcKqjV1/1/9YnT3aBYaKjiNIZxp4KEkt++slNYXnHHanZEpUMM2fCNde4mXuGD3cjW376yQWD558PXbq4jpcRI9wInOnTXfAYb199BX/6kzv/UUcl5hrRUkAoksq8qTqMce9cwV1GlWnb1gWP4Lq5/vGPxJQRSrYQgrqNI/Hii65+wLUO1q7tuo091XUc4caNrtkFKDzgADZ17UqhF8SMGuWSaKqTvDy4/faS60aPhn/9K3HXzM93XwqhbEAYIrEkLw9uvRUOOADeeccV98YbE1e8VLd6NTz0kPtV9e0LDz/sf7+qXx8GDvRvDQ2uxe6dd9wom/33h6ZNXdB2113xmcVr/Hg4/njXsJyZCf/8p+v0SRUpVBQRKWHGDHjrLbd8xhnQs2fk57jwQvehNX68+xA+/XSXiBJP27eXfbf89ltX5hS0Zg20apXsUgTs2gX33uuW99wTzjrLLXsthAC//cbOfoewYoX7NS9f7hphveXGjd0HfzR/Hgk1dWrxYmG/fmAMeZdeSv0LLoD16+G11+Cvf01iASP09NP+3/m//gXPPQdLlrgIrGtXN0wj3hYtckEhlA0IvcSSwATV06fDuee6qUmD3X+/+254xRXxL17MnnnGvTfddVfJv/lSVq6EN990v4oGDdwjOzv0ct268PXXbjKGcePKtpAOHQpnnw0nn+yCwcJC9yv87jv3PXbiRH9Ew/bt8MUX7vHQQ64MRx8d3UudONG1TObmuokhRo+Gk06K7lwJY63VI4wH0AOwc+bMsYmyZcsW+/7779stW7Yk7BoSvSqvn2OOsRasrVXL2kWLoj/Pr79aW7euO1eXLtbm5savjNZaO22aO7dXVrB2n33ie41KhFM3u3ZZe/LJrnhnnmltQUEVFrA8zzzj/+6eftpaa21enrWP3LfTFmKsBftA/duKdynv0aiRtV9/XcF11q2z9v77rZ04sWpel7XW3n57cQG3LF3q6mf9emv32MP/GykqCvt027dbO2mStU8+ae2FF1p70knWvvSStTt2JPA1eLZts7ZFC1fuvfd2f0xz57pfPFhbr577P4jS779bO3u2tYWFpTa8/bZfyVOmlD3wvPOsBbujbhObmVFUvGvv3tZ+8IG1LVu658ZY+9Zboa+dtM+dLVuszcx0BWzd2v0+SykosPbxx61t2LDiv//KHp07W3v33dYuXVp5sYqK3NvtSy9Ze8EF1nbt6p8nI8P9G0XwZ2uttfb7763NzvbP8frr4R8bS/3MmTPHAhboYcOJc8LZSQ8FhFLF9fPtt/670CWXxH6+++/3z3fjjbGfL9grr/jn9iIusPaPP+J7nQpUVjdFRcWfncWPCy6I/I09rnbtsrZDB1eYdu2szcuzhYXWnnaaW7WMdtaCfZWRIT/kWrWydr/9/Oe1a1s7enQ51/rrX/0dBw609pNPQr74hQtdoHX22dauWhXj6/O+0HTpUrJ+7rvPL8tnn4U8dMMGa7/80v3Znn66td26uQ/SUL+H3Xaz9tpr3feehPnXv/wLBv+SP/3UL1ibNi6yi8BPP1l76qklT/G3v7ngPj/flgiq7bZtZY5f+vcnirfvxSJbu7a1d9zh/rSstXbGDD8QqVPH2vHjy5YhaZ87EyaUrMjdd7d2+vTizbNnW3vQQaHrPJxH06bW/t//WTt5cuz/52+9ZW39+v65Tz/d2pyc8I6dMsUPaI1xb5eRUECYgg8FhFJl9VNU5D60wbXsRfghE1J+vrX9+rlzZma6T6J4ufFG/7yffOK/a773XvyuUYnK6sYrovem7C1ffXUSg8Lnn/cL8tRTtqjI2iuv9FdNrDPUWrCLmh9o77jD2hdfdIHCokXW7tzpn+b1110w6B0XsgWjT5+yn5h9+rhPuoICW1Rk7ahRJT/0mjWzduzYKF9bUZGL1MDas84qWT8bN/oXGjbM5udbO3Wqtf/5jwtG27ev/MO+cWNrWzQrKLN+2DBr338/EEzFy4YN7oJgba9eZZvxHnvML8B++7mmzEp+NRMmWHv00RW/xubNrZ3W8RRrwRbtuWeJc+TmWvuPf1h7sJlcfMD1HcbYWbPKXu/zz/2G+8aNrf3555Lbk/a58/DDISt259ff25tv9ssM7nvTp5+61uB166xdssTaOXOs/eEHa7/6ytr//c/F6c8+60773nsl/0fiYeZMa/fcs2RVL1tW8TE//uj/6YC1L7wQ+XUVEKbgQwGhVFn9BAdV110Xv/POnOm/y+63X/w+NU880XotQXb7dv8aV18dn/OHoaK6efRR/9fZqZNreejc2V93++1VVkxffr61e+3lCtC2rbU7d9p//9svU/fu1uaedaEtbjmpxNdf+72XYO1llwV1iRcWui5NcE1tpfrf8vfqYv/T8wVbm7yQgclZZ1m7eXOEr2/hQv8ETzxRon62bbN2+Yl/K97er97cCgOjVq1cY+PNN7sA9bd5ebZo5Ehb1KiR/fHCUXbIkLLHtGvnughXr464Zsr6xz/8E3/4YdntRUWuKcrb56STQvT9ut0+/NDaAQNKljUz0w1hePxxa484wu9FBWvnsI+1YMfVOtaefbbrBv72W/f3AdbWZYfNxx1QcO315b6E4Eb8Pfawdvlyf1vSPnfOOssWR76PP15cwO2mgR3KVxZcy+l111UaY1eZP/6wJf7eWrRw9RHKjBmuldLb95lnorumAsIUfCgglCqpn8JCa/v2df+aDRtau359fM9/003+O9R998XnnF26uPOdeKJ7fuCB7vn++4fcfcoU1xI2ZkzIXrColFc3b7zhtwi2bOkPxVy2zAUN3q/i4YfjU46wvfSSf/HHH7cvvug/bds28IF9773+yjD+5mbPdsd6h/z5z4HxdYsX+ytHjXItdHfd5ZoAgyKTZbSztzZ5zH7zcY4dM6bkh1n79tZ+800Er+/VV4sPzp04zb7+eo494YRf7X77FdjMTGv3ZmHxGMmn+Wvxddq0caMO/vUv972oTEBXUOD3qXuPl1+2c+e6ILj0WLNatVyXbNRv26tW+cH0wQeX35y8a5eL5rwL33RT8ab8fNeK27NnybLVrWvtpZe66gm2fr1rSTrxmDy7C/fl6j7+HjJY3m8/a3d06eWeHH54hS8l+AtHjx7Wbtrk1iftc6dHD1eYYcPshg3WPj/wBVuA6zvPJcte0emj4B7kyhUVWfvxx9b+6U/WXnVVyKA8Hnbtsvbyy0v+jY0aVXKfWbNK/ns99VT011NAmIIPBYRSJfUzZoz/LnLHHfE/f26uP0q6bl3XkhOLnTv9Jg1vbOLf/26Lmz62bi2x+4QJ/uerV4QTT3QtGN4HVDRC1c0XX/hdqQ0bum/swRYs8PMEwPXgVon8fL+JsnVr+8m7ucW/wiZNgoKXt97yC1e68OVYscLafff1DxswwNotoz/yVwQGkW3ebO0Fp223V/KwXUFQFOm12Dz3nF250nXBequNsfaaa8LLSSq89DJrweZlZNmmDUK3PL7PCdaC3ZWZZcc8sc4uXlxJ931RkbUXX1z2RJmZrp/Yui8YTz/tenaDd2ncOMoxhn/7W5nfXbk2bSqRgZD/wit21ChrO3YsWZZGjVyj45o1lVx7zpzigx7d76US/zd16rigedcu6w+ObdKkwl9gUZELmr1zHHqoq8ukfO7k5BQPnJx74o3F/4d/4c3iILioVq3yM2GCFRS4983evUv+oj/5JKEv4bnnSg7VuPhilxA2Z45r1PfWP/ZYbNdRQJiCDwWEkvD6yc/3W9t2371MMBU3Eyf671axJqz8/LN/Lm+09P/+568bN65410mT/AHuoR61a7uuweeeizwfpXTdTJ/uX6t2bTfOKJSZM93nKLjPp3A+f4IVFLhuwGefjaC1M6j1bMlVjxR/0Neta+133wXtN2OG/8uJoGCbNpXs1rqv+QP+k3Xr7Ndfl2wdbdV0p5128bOuPz24QmbMsEVFLqs3OBjp0SP0ENSiIjem6/LLrZ1R6wBrwX7HgKC4rdDuv3++vfZaF79tfv8b/6R33ln5CwseCLrvvq7/1CtYnTolKrmoyGV2nnGGf0jPnhF2PS5e7H/iH3VUeMf8+mvx2Mk8U8cO4Lvi67do4Rp9w+5+D/5yOG2a3b7dJR3feGOpFs8n/MSSymYjKCgomfd1yinWbtpU/vtaYaFr7XroIWuHD9tlT2/xhR154jb77ruxjdHb/Jk/9vFk3i4uz9FHW7v62f9Zm5Xl/1O+9FLok+TluTeL4PEfwY/zzou+gGGaNMkNafAueX6vafbyRi/Zs3GPz05/yZW/vMekSZVeQwFhCj4UEErC6+e55/x3lv/8JzHX8Awe7K7Tt29s5wluxfKm3diwwV93yy3W2pKDq41xXWIffmjtueeW7JoMbvQ57DAXjIQzDiy4bhYt8lv+jLH2zTcrPnbSJGsbNLDFwWM4DQtr1rgxasGBVfPmrtoqbEErKChuRcpv1tK2abqj+HPvgw/KvCj/5PfeW3mhguzc6fesPsf51oLd1biZvfrqkr/no4+2duXKwEH5+da+9pq/8eabi883f761/fv7m2rXtvaee9zLmT/f2ltv9ePJLHJtHi6QeiTzajtihLWvv55j33zzw5L/O0VFfrJLy5YVRxjBWfJ77eWnQH/yiT9mtUGDkFOzeA3W4LJDw04iOvts/8AffwzrkA0brH3guG+KW7nW0twe0naJffLJKKbHue02//oVRbKT/eDKjhlT6Wlzc60dNMg/5P/+b6d97z3/fW3ZMtdafvrpJVvQX2WktWBf4mwL7v/5ggvc+NVwpnDKyXFDOIYPt/ayjCeLT9yR32zz5i4ppLhuvviiZIbTk0/6J9q+3dpHHvGnL/Ieu+/u/im9Zu0mTVzQmGC//27tAQdYeyCTi7u8w35cfHGl51dAmIIPBYSS0PrJzfWjiz32iP9cgaVdc43/yR7LV/077vDf3IKbyLwBU4MH21mz/IRTcHFvsF27XEPiRRe5oKr0e6YxrsVr1CiXYRiKVze//rq1OFcDwu+u+fJL18gErqVuwoSy+xQVuQHkp51Wsquo9KNtW1fWkJ9Fo0cX73hHk4eKj3n22XIK5vU9XXhheC8kSGGhC4Ym4ebu+JaBxderX9+VMWRwtP/+bqcePUqszs931R2c9BAcMHiP4MzXnBdcNF7u/87LL/sHltcSFDxXY+vW1v72W8ntb77pDxTdbbcyAwbz892XC+8UYY0XnTPHP+fJJ1e6e1GRayD3/n7Px/9yV9h3v+hS2b2mvI4dK95vxw6/Uq4vP7Ek2MaNbhpI73cyfPgie8EFeeU2tnVlXvGYz800sobCMn/z117rGrWDX2p+vssOPvNM/0sXWPssF7hzmcb2huuLQg+V/u67kplSd93lAr7g/ljv/fLRR/2gOeh/zH78cXi/6xjl7iiyv7Q8tPw3BQWENeuhgFASWj+PPBL0blledBBHwW+aMUyoa0891Z2jffuS6y+91H0Y1smybXffGfKLfigFBW6o1uWXl0yQ8B6ZmdYeeaQLKjds8I/bsmWLfeONj2zv3v5UJJFOt/j++/7nasOG/q9lyxZXbm8MfPCjVy83Zm30aL+333t06OCmiilO5i4sLE4PXZ/Z3NZne/HnXLm8BJ2hQyN7MZ6iIruzXmNrwf6Xiyy4lr4FCyo45q67/BcRYuDd1KklJ+v1Hvvv74KtzXcG/S0vWWKtreB/Jy/P73Pr3bts4DRmjB+YNW1ads4Uz9NP+9ds06ZMpsa6df73rczMyocD2j//2e2ckWHtL79UuOv8+SUDTnA9zJvOviK2/7Fu3WwgWqt8X2/QZCWJJcGWLXO/qvJilaZNXUw6apS1W0acX2Ljpw/8bE84IfQXo27d3AiAyy8P/YWhSRNrlzVziXNFlf1d//hjmeSn4kfnzq4ps/Q3r61b/Yn4zzkn7N9HTMaNKy7XsuP/5v7+wnmEMTZGAWEKPhQQprlFi+zOSy+1395zT/zrZ+tWv2mhc2d/VtkohZVct2CB/8ZaOkUuEt4H0bBhJde/+Wbx+b1xVJH2ghcWuu7cK68M/cFVq5Ybc/jSS9b+9ttW26vXuuJt550XXaPMa6/58UezZm4+59LjHuvUsXbkSDdGrXRryIsv+nNNe48uXVxXWeEb/piw67jfgstZqLCcI103nW3XLvIXY63rWg1c85GOD9t77gljtqHgcaEPPhhyl5wc1yJ0wAHW/vOfpQJMr6+6RYviF1fhe1twABo82PPTT/2Io0EDN0CxIsFptMHdygHTpvlD01q0qGB6z6lT/fOce265l8vNdb26XssyuJ7vN94IvOwVK/wNkX47CU7WuuGGyvcPM7GktNmzrW3c2N3hJCuryB5+uBudMG1aUDfwihVlI7/AXXU2bHCL3giU8h5167rxiu+/b+3OLTv98117beWF/PnnkgP1evd2XxQq6qc+6SS3b+PG8Z+QsLSiIn+O1+zsuE/Ir4AwBR8KCNPY6tXFzQsFtWrZ7YGMxrgJ/kB8442oTlFQ4HrW2rd3b74HHOC6YEeNcp+jZcYvFRb6c3RE0R1ZfFHvE7bUnIPLJq8sfk3/4B57zz3RXSK4uN9+6wIo73Zc5T2GD49tisVRo0Kft0MHF3OU123tyctz5ygdxE6v7yYbX8futgHb7IgRYYy98saRGRPdB9uXX/oFKOeuIGUUFfkDAgcOjPyaXlrt8ccXr6rwve2PP/wWHa817LvvSiaMfPlleNe+4Qb/9e67b8lmZFtymO5BB5XzKz3ySLdD7drFLZylffFFyVwGY1yjeJlMea+Ft0uXyL6hzJ7tnzycW1tEkFhS2vz5W+0993xr16wp53Pn2mv9c3vRb4iWt+XL3UxW3nfEjAz3q3zppVKzJk2f7p8v3Hu4LVnixrR+/HF4v8egL6Qh546Mp7Fj/Wv9859xP70CwhR8KCBMUzt2+G/qgUdRgwZuIHc8bNjgj5Pp3TuqubPGjSs7x1npR0aGGzM0cqRr9PnqK2t3HXKo2xhtYsmvv/oXCJp1dflyFxP8igsqFux9THTnL0dBgSv/xReXHU7Uv39+2LeUqoh3dzVjrD32WGs/+ijyex/v2OFaRZs3t7YNvxcX8n6us0OGhDlMNHiM3bx5kb+QoAl/S8xGXBlvjGlGhrVr14Z/3Lp1/vXuvrt4daXvbRdd5B/31lt+BlJGRmR3vCkqKnmuAw8sk/4dvLlMkv3XX/sbL7uszOkXLXItXcF/c336hL7NsLW25G36yuvuDuWNN/zjwpmML8LEkmAV1s3GjX4T+ZAh/u0IO3eu8JxLl1bwxenZZ2P7mw7Htm3+F4qzzkrMNax1bwpe136zZmHNFxopBYQp+FBAmIaKilyqXeDNK3/AAFsUPJ5p9uzYr3H99f6b40cfRXTo3Ln++7P3aNfO3TRh0KDKbwj/IO5DPz+jtn3m8Z125swIg57g6WUmTrTWul46r+XkBc61FmxRo0aRR1Nh8hJSzjknzx5++FK7ZEn8puqZPbvyW1OFY9s2az8/7pHi39XpnaeFP/XId99F/fdhrS0ey2mzsyNroQq+l3YkEzR++KF/XFCrXqXvbb/8EvqPtLxEk4oUFPhjW8GNqwtqCty5s2TGdPHtxIqK/Jvn1q9fIr193To3Ji74dmoNGriAv8LW6EWL/AMiuSXOLbfY4m8k4XzDiSKxxFNh3dx9t1/+Tz8t2ZtRWVN5eby7ujRokLD3BWuttSNGuOs0apS4JL3gCebLGV4RKwWEKfhQQJiGgjNoDz3UbvnjD/uT9wELblxLhN0zJaxc6X+LHTAg7A/stWtdy0Zwtmd2tpsGJLhruLDQNeKNGeMmwh02rOQg79PwE0v6Ma34PIcd5m608OGHlQyHCR6ztX69XbvWv6UWWPv6ES/4T8KcWDlaKf+/E7hfWU6rjjZnewSB2erV/u/wkUciv643IWE5d40pV0GB3/wa1PVbqeBAJqguwqqf0jf3jeb1evLySn5batHC3Wnk1FOtvf56u/GuJ+wZjT60PZllm9fZ7GaVCf6CExjzl5PjJoAu/eVq5MgIGly9CZN79Qq//F5SS6dO4R8TRWKJtRXUzY4d/thmL+Hnq6/8X0KZeZLC5PW4HHJIdMeHK3hKrGjLWpGdO/2bG7dpE8W8QuFRQJiCDwWEaSZ4UthOnaxdv764fnLvvNPf1rFj0ERuEdi5s+SsuaHmOSklN9cN+A7+cMrIcF2nld71IKCoyBX3o4+sfewyP7HkYkaV25LYqZMbk9i3r+ua7t7dDYkam32OtWDXZbSwLVqUnDbs4outLVoY1KUcy4d7GFL6f2f5cv/3EE6CQLCiIn++jssvj/za3oDLs8+O/FgvUaFu3fBndPbG3+2zT4nVYdVP8HjHeIzFyslxYyAraiYPPLaYRrYwO/CP1aSJLfhjo33uubLjQA8/PLwe3BKC3y/CvV2Kl7J+wgnhXyfKxJJy6+ZJf65AO3q0W7dtW/EdRiJtibTWuuZUb7xoNH/Pkdi+3X9TGjky/ucPHo7x3//G//wBVRkQ1kJESpo6Fc45xy03bgwffQTNmsHWrQDsuvJK6ubmwr33wpIlcOSR8O23bp9wfPopu/7vCuosWwTAzFbDePzlwWS/Aw0auEd2tr/coAGsXw933gnLlvmnGTYMHnwQ9t03/JdmDLRp4x4csze83BC2bePeEdPpfwz88ANMngxz57p3OoDffnOP0vZgHgBzivZh3Tp//bnnwlNPgTGdoFUrWLMGJk6EK68Mv6A1ydix/vJf/hLZscZAp04we3boSqjIxo2wdq1b7t49smMB/vQnePFF2LkTPv8c/vznivcvKnL/OwAHHhj59Q4/HN55BwoK4JRTIj++tPr14dNP4emn3e9v+XL3D7RihbtGkEZ2K2x3y/NPuJ4RQ5oyd66/vVcvuP9+OOooVyUROekkuO02t/zOO3DDDRXvv3MnLHLvDfToEf51+vVz9bV5Myxe7P5uolVQAA884JY7dvTrIzsbeveGn36CSZMiP+/8+e71Aey3X/TlC0eDBjB8OLz1FnzwAeTmQr168Tl3Tg7cfbdb7tQJzj8/PudNMgWEIsFWrIATT3RvWpmZ8Pbb0K1b2f3+9S/3xjtqFPzyCxx7LHz5JTRsWO6p101ZwrYLr6bTnA+oE1g3hx78ec0olr4QfhF79HCB4NFHR/TKysrIcG/KEybQ9LfpnH++/762dav7bJ88GaZPh7w8qFUr6JFp6fXuPMgHunXn4kPd+j594Lzz3KnBwODB7g154kQXYUb8aVoDvPWW+9mpE/TtG/nx0QaE8+b5y9EEhEce6YKqHTvcB2plAeHChbBli1uOJiAEFzzFU3Y2XHttyXWFhe5LyvLl2KXLGP3v5WyevYw9Wcbmhu248JWryQvs2q4d3HUXnHmmezuIyj77QNeusGABvPtu5QHhggUuuIbIA0LP9OmxBYRvvw1Ll7rl665z/9yeQw5xAeG0abBrF9SpE/IUIc2Y4S8nOiAE9wXsrbdg+3YYN859yYmHxx7zv2zdeSfUrh2f8yaZAkIRz/btcMIJ7sMC3D/9kUeG3tcYeOIJFxS+8YaLnv70J/j4Y6hbt3i3LVvggzdzsfffz18W/5sWuG/HW2jEP7mDcZ3+Rp3M2rTZ7r505uSUabwo1qKF+3A6//yS788x6dcPJkyAOXNc1JeVBUCjRnDEEe4R0u8rYcw2AIZe2p2hl5ez36BB7g153ToXMHTtGqeCVxPLlrlmV3AfTtEExN4H++LFLpgJNzIJDgj32Sfy69ar55rE3n8fPvzQ/WFW9IfnvU6IPiCsCpmZ0LYttG2LOfhg/nQCHHSQ+xfA/UnTuDHcdBNcfnkcGpWMgZNPhnvuce8TK1a4SLM8wU2TkQSEvXu711ZY6ALCSFujPdbCffe55ebN3Te8YAMGuPe+vDwXGEZS115AmJUV3ZeUSB1zjGspzMlx70PxCAg3bXLNxQA9e8Jpp8V+zhSRkewCiKSEoiI46yyYOdM9v+wyuPTS4s2TJsExx9TnllsGcNZZ9bj4Yrjplgwe7vMyK3of53b6+mu2DT+dNb8X8N57MOJky/m7/4+Bl/TgnMW3Uy8QDL6bfTZPX72AS+ZdxbxFtVmwAFaudLFlfr57n92wwfVwzZsHP/7orr94MVx0URyDQYD993c/8/Ph55/DPy7cYGPwYH954sTIylYTxNJd7Nl7b/czPx9+/z3847w6qlPHdftFw/sA3bgRvvuu4n2nTHE/69ePbBxDkjVoAO+9B61bu1/V1Ve7xtjrr49fD2OJls933614Xy8gzMgI3TtRnnr1/ABy+vTIyhds3DiYNcstX3ll2V/CgAH+cqTdxl5A2KtX1bSq1a8Pxx/vlv/3P9dtHKsHHnBv1uB6ijJqUBgVzkBDPZRUUuMFT2g7bFiJuSSWLCn/7klgbV122PH4U/W/xNm2Mwvsx5ScE2bF7n3snKe/i+oOGgkT7R1Lgm+1V1FSTWGhG+QOCZ0PLO7/OzNnWnvccS5rNpYK8+Y36dw5+vN88YX/uw6+k0dlvKzdffeN7rrWujRzL4ngqqsq3revux2ZHTSozKbq8N62fXv4uTMRKyryb2EzeHDF+554ov83E6koEkvK1M2hh7pzZGe7eQhLKyrys21GjAi/bIWF/pyGYdzDN27efdf//3nnndjOtXq1n6hy0EGxvTeEqSqTSmpQaCsSpZde8rtIuneHMWOKm+F27HBDpzZscJs7d95Ely6FNG/u99ztpB7H8yHTcWNizuEV5tONY/kUgPzsJhQ+9iR7rPmRHhcdklrD6Pbe2x/3GEmrgtf61KiRa1opT0YGDBzoliNtIczNhYcecs03XoZLohUVwcMPQ//+rvv/7rth/PjozrV0qZ9kEW13MZQcCxbJOEKvjqLpLvbsvrvr9gfXdVxePezY4cY5gut/rYa8BK6EMMZvJZw40R9/ForXQhhJd7HHG0foJZZEasoUN4QE4OKLoWnTsvsY47cSTpoU/v/mokVuWA5UzfhBz9FHu7Gk4I/njdY997i/dW85pd7MY6eAUNLbxImuHxZclvCHH7oBRLj3uYsu8nuRL700jwce+JZp03JYt86Np960yb3PfTmlEZvf+Iwtrd0YuQws1hi48EJqL15I5uWXxjAqPYG8xBKILiDcZ5/K3xS9gGLpUjd+Klx//asb0H7SSTBkSGRd2tFYs8YlB11zjatcz513Rne+t9/2l6PtLgY33swbJ+Bln1YmJ8dPSY91rJbXbbx0qR/0lTZjhhu7Bqk9fjCZvIDQWhdch5Kb6wf9sQSEEF23sffFuHZt13deHi8gXLXKjW0JR3B5qjIgrFfPjQ0H9/7uBXSRWroU/vtft3zkkTB0aFyKl0oUEEr68sYN5ue7N8B33y3RGvPoo/D66255yBC48868EodnZECTJu6Q/v3h8NOa03jKF3DooXDEEZgffoBnn3UDs1OZ9yHiJZaE45df3M9wgg0vIITwWwnHjPF/+eCm9enbF666yh+/E08ffeQGiI8b55537Ohn1Y4f764fKa81omtXd+5o1arljwEMt4Vw/nx/OZYWQnBZ954PPgi9jzd+EBQQlufgg/3W9PLGEc6f77e4RRMQeoklEHFAmLFwoR+onnWWS7wpTzTjCL3xg7VqVf0YU+8L2Y4d8Mkn0Z3jjjvcZwW4sYM1kAJCSV+rVvmtKLfeWiIB4ptvXOMUuAaaMWPCHAPdrp0LIL74wkWJ1YEXEIabWLJ+vXtAeAFhv37+wPRwAsLff4dLLnHLzZrB3/7mZ08++qgLsF5+2Z+aIxa5ue78xx/vv6Yzz3TNwk895WeMR9pKuHixywaC2LqLPd4XlXADwlinnAnWsaNLAoDyW7a8gLBNG9hjj9iuV1NlZPhfMr7+2nUvlBZthrEnhsSSOo8+6k8N9fe/V7xz377FMxLw/ffhXcALCHv0KDETQ5UYNswfGhNNt/Evv8Arr7jlP/8ZDjggfmVLIQoIJX0tXOgvB7VqLF/uPsMLC9173nvvuSlfaiwv0xj8IKYikU5nUqeOP66sspa2oiI3s7XXCvjss26Ki59+ci2v4KawOfdcNzYxeF6zSM2e7V77U0+5540auVbJV191y61auXFUAF99Ff4HH8Svu9gTHBCGM2bLq6OMDOjSJfbre93GP/0UuovQm3JGrYMVO/lk97OgwGW9luYFhJmZ0U/R5H3Bmz497PF9ddevp/aYMe7Jn/5UeXZznTp+UBROC6G1/v9qVXYXe+rW9Vu6P/rIDamIxG23ufcmY/wJqWsgBYSSvn791V8OfGjm5rovgF5j0TPPlByWUyNFmljidRdD+K1PXuvrL7/4v9xQHn3UBV/g5j/zWlR69nTNtm+8EbjNCm7W7P33d9MDbdwYXjnAvbE/8oj7QPNeyyGHuKk2zjij5L7XX++3hETSSui1QnTvHl1LT2leQLhtG/zxR+X7e6+rUye//LGoqNt49Wp/bKgCwooNHuzf0ShUt7EXEO69d/T1FkViSacPP8R43aGVTZzt8bqNZ83yk0XKs3Sp/yUvGQEh+F/McnNdwli4pkxxd5gB15Ue6xCMFKaAUNKX10KYlQXt2mGt66n0vshefjmcfXbyildlIk0s8Vqf6taFPfcM7xrB4wjLm89uzhy48Ua33LGjCw6DGeMmgZ0/3wVqtWq5lodRo8ju148ub79N7WefdS1+FT2OPdYNmN+1y73222933fwdOpQtU5s2LrkF3O3bgidfLs+iRf4fUTy6i8GfixDC6zb26ihek//27etPply621jjB8NXq5YfXI8b5wL8YLFkGHsiTSzZtIkO3tjZIUPCr8NDDnE/g29ZWJ6qvkNJKEcd5Vr+Ifxu4y+/9G8JVbu2e6+owRQQSvryAsK994bMTJ54wh8mMmiQm/EkbUSSWOIFG926hZ85fdBBfqZsqHGEeXkwcqT7mZHhum3Luw1gw4YuG/Lnn4vvJJOxcSPdX3+detdd58YEVvTwPvw6dHBl+ec/K57t+4Yb/NtzhdNKGNxdHI978kJkU8/s2uVnI8crIDTG7zaeMKHk+DcvIMzIKDn8QELzuo3z8komOOzY4e6NDrEFhJEklhQUkDVqFLW8+wuH2zoILknGU1m3sRcQGuPKlwxZWX4w/vHHlbdqPvWUCwa9ls177ol+gvdqQgGhpC8vIOzcmQkT/FkW9tjDfabXkNtThieSxJJIMow9DRr41wgVEN56qz+lyY03+q0PFenWzQV377xDUfv24ZfFGP+uNMHZkuXZYw+44AK3/OmnlbeGeK0PPXrEp7sYSn4QVTb1zK+/+lPAxLN7y/swLSws2eXmBYT77uvP9yblO/xwv6UquNt43rzYMow9wYklP/zg/l8/+wyefhpuvtklTQ0e7Fr369YlKzDVTGHPni75IlzNm0Pnzm453ICwW7cETvYYBq/beOdON5YwlIICd6eqv/3NH0j++ut+lmENpnsZS3oqKCgeX7O1VRdOOcX/33/3XWjZMsnlq2qlE0vKa+nZts2/fVqkrU+DB7vgYcYM9+3cCx4mTIAHH3TL/fq5FrtwBSb83X7YYXz/wQcMHDiQhuW1LHrq1vU/kMP1j3/Ac8+5gPmuu9x8ZqEsXOhPXBmPZBJPvXpuGpCVKytvIYxnhnGwwYPdPEubN7txhGee6f5ppk1z29VdHJ6sLBg+HEaPdoF1bq6r31gzjIP16+e+YH37bdjnyrv+eupHOrxhwAD3BWTyZNd1HOo2btb6LZXJ6i72HHmkm2d2yxb3xa30fYg3bXL/t19+6Z63auWGSKTJ37ZaCCU9LV3qgkLgsc+6FI/THzWqxs4oULFwE0timd/OG0dYWOg+QMC9MZ99tvvQqFcPXnstuqbZjAx2NWmCbd7cpYRX9Ig0GARo394luYBrWSgvuzkR3cUebxxhZQFhcNJPJPfCrUzt2nBc4L7dn37qWlnmzfO73tLkQzMuvG7jnBw3NhX8gLBWrdgzw72M/NIyM93f8sCBbojGjTeS+5//8O2//02BN3lzJLwW9s2bS743BFu1yk+ESnZAmJXlD3345JOSYzgXLnRDW7xgsG9f1xuQRn/XCgglPQVNOfP5Utftceml/md+2gk3sSSaDGNPcDewN/3MZZf505g88EB8A5h4u/FGf6xheWMJve7inj3j2zoH4c9F6LUQtmtX/jjMaHkfpjk5LhtcCSXRGTbMn5vTy2D1AsLOnf0xq9E680x48kk3gfJrr7n/t2XLXBC/bJkbtvHaa3DPPeRfcAGbov2/C2eC6lRIKAnmtdzn5fkt/V995f5+vc+Fk092vyMvkSpNKCCU5Nu2LfzbH8XJ+sn+lDML6cIhh7hb2Ka1cBJLvGCjVq2Sma/h2G03/44dEye62b5fe809P/poF5Gnsg4d4Jxz3PIHH/hdw5758/1xkPHsLvZ4AeG6dWWzU4PFO8M42LBh/nQoH3zgZ11nZyfmejVVgwZwzDFu+cMPXSJQPDKMPZmZ7v/ppptcS+CgQa5lsKLkqWjss4/f4l7ePJ3BAWGfPvG9fjSOOMINfQD3BW7UKPd37SWP3HKLW5/MsY5JooBQkis/3zXTd+rkBj5XgXfegffuc98Et9KQDv1b8s47sX8pr/bCSSzxWgj33ju6X5jXbfzDDyXvRvLCC9XjRvE33eRncN51V8ltiewuhvAyjQsL/a67RARoDRu6pAhwAaHX9X/AAal5r+5U5nUbb97shiEsXeqexysRqSpkZPjZxpW1EHbq5AdiyVSnjj+/6QcfuMA5OHnkrrtCj4VMA+n5qiV1zJnjgoyCApcBF+bM+tHIy4MrroARI6BDvgsIt7TswrcTTfolkYQSPH9ZeXcsibX1yZugOi+v5N1IvHu8prq99nIZyuCyj4IDZ6+7uHfv6O8yUZFw5iJcutRv3U3UBLpet/G6dX6rlrqLI3fccf542eC7X1SngBD8buOFC0NPOp/MO5SUp3QLfqtWLrmt9MT0aUYBoSRX8Hi1GTP8u1TE2eLFbhz144+7592MCwjbDe2slkFP584VJ5bs3Onf+SDaYCN4gmooeTeS6uKmm/wWBK+V8Jdf3JcbSEx3MYTXQpioDONgxx9ftjXXuzWhhK9x4+J5NPnpJ399dQ0IwW8x9qxb589KkEoB4eGH+19C0zB5pDwKCCW5Sgce//533C/x7rvuvchr9Bq0fy57ELjVVjzu81pTVJZYsnChm1oCog822rTxf+eh7kZSHXTu7MZlAYwd61rJEt1dDK67bbfd3HJ5cxHGkvQTrlatygaA+jCNjtdt7Kld25/br7o48ED/C1LpbuPgQDeVAsLatV3jwwsvpGXySHkUEEpyle6a/Oqr8rsrI7RrF1x5pXvP3bLFrbvqKvjqmd8wXte0AsKSKkosiVfr0/PPu4meP/kk/lmwVeXmm92HoLWuu8/rLu7bN7Ef6JVlGnt11Lw57L574srhdRuDS1Zo1Spx16rJTjih5NjLLl2q34z4DRtCr15uuXRAGJxQ0rdv1ZUpHN27ux6KNEweKY8CQkmeXbv8rMyTTvIz4AIz58di6VLXRfzYY+5548aupfDhh6H2En/KGQWEpVSUWOIFG8bENj3MwIFukudUnmKmMl27+pPavvmm3zKXqO5iT2VzESYywzhYcECo1sHo7b57yTkDq1t3scfrNp461b13eLyAsF079yVFUpoCQkmeuXNdUAhuHJk3oPedd9zs91H68kv3ZdS7gcL++7uei+KhakFzEFa77plEqyixxAt69twT6tevujKlqltuKTuWLlHdxR6vhXDFirItuNZGd1vBaHTp4qbvgLJ3e5DIBHcbV/eAcOfOktMxpWJCiZRLAaEkT/A4tX794Prr3bK1bpLiKDz3nJvey0tgvfxy+O67Uvck94LN5s1TYxqEVFJRYklVtT5VF927l2wR7NevZOJHInjnLyrypynxrFrlz0+YqAzjYJ99BmvXutZ9id5JJ/ndll4WfnUTnFjizUe4aZOfhKaAsFpQQCjJ4wUcDRq4FocePVwGI8DLL8Pq1WGfqqgIbrgB/vpXN4ONN6XUY4/58+gW81oI1V1cVnmJJQUF/u+tKoKN6iK4lbAqWsoqyjSuigzjYJmZ7laAEptWrVx27hdfwJAhyS5NdDp08MeReuMIg1sKFRBWCwoIJXm8Lsm+ff2B1Tfc4H7u2hV2BuqOHa6h5v773fPdd4evv65gSikvsFF3cWihEksWL/a799VC6Nt3X5dpfMcdLoMp0Sqai7AqMowlMXr29LvgqyNj/FbC7793vTypdss6qZQCQkmO4ISS/ff31x9yiEs6AHdLIS89uBxr1sDQof7tQLt2dTfBCO7BKGHLFjc3FqiFsDyhEkuquvWpOjnpJLjttqrJDm3Vyh+/WV4LYcOG0LZt4ssiEsx70121yo1x9QLCli2rz8TzaU4BoSRHcEJJcCID+K2EW7fCf/9b4SkOOsgltoELDCdPrmQYV3CyigLC0EIlliggTA3GuLulQNm5CIPHeFaH2wBKzRL8LXzSpJIJJfp7rBYUEEpylE4oCXbssa4rDtw8MTt3ljn8iy/c+8+yZe75eee5Me5Nm1Zy3YWacqZSoRJLvO7IVq3C+CVLQpU3F2FVZRiLhLLffv79zT//HBYs8NdLtaCAUJKjdEJJsIwMP+N47Vp45ZUSm595xmUSb93qnt9zj5vrOKxb0AUHhInOCK2uQiWWKMM4dXjjCJcs8e8cs2ED/PGHW1bSjyRDVpY//OfNN/370isgrDYUEEpyeIFGcEJJsNNOc3dAADcFTWEhRUVw3XVw8cVQWOjef8aMgRtvjKBHwusybtdOc+lVJDixZOdOPyBUsJF83heZvDxYudItq0tfUsEhh7ifubn+OgWE1UZKB4TGmCxjzH3GmFXGmFxjzBRjzJERHH+qMWayMSbHGLPZGDPJGHNYIsssYdi1C2bNcsvBCSXBateGa691y4sWkfPau/zpT/DQQ25V8+YwfnwUN4bQlDPhCU4s+eQTyMlxzxVsJF9wy7Y3jlAZxpIKSmfzNW3qJrKXaiGlA0LgJeAa4HXgSqAQ+MQYM7CyA40xtwNvACsC57gFmA0o/S7ZKkooCXbBBdCsGQBLLv43H37ouiC6d4cpU1xCSUSs1ZQz4Qqul1df9ZfVQph8oeYi9FoIs7JKzcIuUoUOPrjkcyWUVCu1kl2A8hhj+gOnAX+31j4YWPcKMAe4HyhvYhGMMQcBtwHXWmsfroLiSiQqSigJ1qAB8464nO5jbmffvBkczlfUP/4IXn3V3Zs4Yn/84Q88VAthxbzEkm3b4OOP/fVqfUq+Pfd09/0uKCgbEHbtGnoIhkhVaNnSfWHx/i7VXVytpHIL4Qhci+Az3gpr7U7geeBgY0y7Co69ClgDPGqc7EQWVCJUUUJJQGGhuwnEwDGXkYMb6/f0Xvfx/vtRBoOgDONIBCeWeDerb9LEveFLctWq5XfDeR+8XpexWnAl2YK7jRUQVisp20II9AUWWmu3llofmHWOPrju4FAOByYBV+C6ipsZY9YA/7LWPlHZhY0xLYDmpVZ3Ati+fTtbt5YuUnzkBMZpeT9rqgZTp5IJFPTqxY4Qr3XTJvjrX+vxxRe1gWa8XOdCLt31GJ0Wf8n2iRMo6ts3quvWnjWLeoHl7W3aUBRhPaZL/XiyevYka8KE4ucFXbuyw7tXbopJt7qpv+ee1PrtNwoXLCBn1SoarXBvhTv32otdCXp/ikW61U91Eu+6qb3//tQLDDPZ1rUrNgX/HquTWOpn+/btEe2fygFhayDUzWy9dW1CHWSMaQrsDhwCHAbcASwHzgMeN8bkW2ufruTalwL/DLVh6tSprFmzpvLSx2CqN9NyDWTy8zkucIeSZbvvzpxvvimxfdmyhtx7b3/WrHF3fdhjj21kX3wgRbdnklFYyJabbuJHb0qaCO3z9dd0BooyMvh68WLs8uVRnacm10+wtrVrE5zys7JhQ2aWqq9Uky5106tOHToCRYsW8dPo0RwaWP9zfj6rUriO0qV+qqN41U1G69bsO2wYOa1b89vy5RDl+6yUFE39LI/wdx9TQGiMOdBaOyWWc1SgHpAXYv3OoO2heN3DzYDTrLVjAIwxY4GfcS2GlQWETwFvl1rXCfigf//+dE/QOKqcnBymTp1K//79adCgQUKukWwZs2aRWVAAQJvjj6f50KHF2957rxY33VSPnBw3CHn48HxGjbI0ajScgl/+Qp033qDN5Mkc3r49RVHMIVjv+ecBsB07MuTIsJPVi6VD/QTL2GMPNzF4QMshQxgaVF+pJN3qps7cufDZZ9TesYMDvS59oPvJJ9M1Bcd5plv9VCcJqZthw2gGtI/P2dJaLPUzL3g6qjDE2kI42RizCHgVeN1auzjG8wXLBbJCrK8btL284wDygbHeSmttkTFmDHCHMaa9tbbc0Nlauw5YF7zOBDKlsrOzadSoUVgvIFoNGjRI+DWSxpu9Hqg3cCD1GjWisBBuvhnuu8+tNwbuugtuvLE2GRmB+8PefDO88QbGWrKffrrCW9qVa8kSADK7dYvp91uj6ydY375+YglQd7/9qJvirztt6qZHj+LFul984RYyM8nu2zfMGdqTI23qpxpS3aS2aOonOzuy9IlYk0rOBH4FbgV+NcZ8b4y5xBizW4znBdc1HOqO2N66VeUctxHXirjBWltYapsX5OneW8lSKqEkLw9OOMEPBhs3ho8+cvFfRvBfZ48e7vYkAO+849+hIVxFRf6k1JpyJjzBiSWgDONUEtxCPnGivy6Fg0ERSW0xBYTW2tHW2uNw4/muBAyuu3WVMeZ9Y8wIY0y071AzgS7GmNIh8YFB20OVqSiwrXmIa3vjDv+IskwSq6A7lBSZTM4+2817DC7m+/FHdyvjkP78Z/dz/Xr/xunhWrHC3dkBlGEcCW9esaZN/TvHSPLttZe/HBiCoQxjEYlFXKadsdaut9Y+Ya0dAHQG/gV0A8YAa4wxz4QzmXQpY4FM4CJvhTEmC5ccMsVauyKwrr0xplupY8cEjj0n6Ni6wEjgF2ttea2LkkhBdyix/fbnqqvgrbfcpgED4Icf/Nu0hjRsmL/82WeRXdtrHQQFhJG49lq48kp4/fVSTbaSVPXrQ5tSeXVqwRWRGCTiHT4X2IHrtjWABU4EJhhjphljwvoaG0hWeRu41xhzvzHmIuBroAMQnGb6ClB65OTTwFzgSWPMA8aYy4FvgT2B66J9YRKjoDuUfLiqH48/7lbvsw98+CFUOtyhfXu/FWTcuMiurTkIo7P77vDII353vaSO0olVCghFJAZxCQiNMQ2NMecZY74ElgH3AEtxk0u3wnXVngq0AF6M4NRnA48AZwGPAbWB4dbabys6yFqbi5tyZjRwPvAAUAQcZ639NILrSzwF3aHkhrfdHUr22MM19u0W7qhTr5Vw8mTYvDn8a3sBYd260FZ3L5QaoHRAqC5jEYlBTAGhMeZEY8xbwFrcHUQa4u4S0sZa+ydr7bvW2nxrbaG1dixwN27C6bBYa3daa/9urW1tra1rre1vrR1Xap8h1toyN0u01q6z1p5rrW0WOPag0sdKFQsEhNtpwEK60LSpa+hrV9E9Z0o7+mj3s7AQvvoq/OOCE0rU9Sk1QemAsFvpkTMiIuGL9ZPxPVySx8NAd2vtgdbaJ621G8rZfxbweozXlGpq2wQXEP5EX7LqZfLRR1E0agweDPUCU1BG0m3stRCqu1hqiuABt+3bu8x9EZEoxToP4WHW2vHh7mytnYp/6zlJI7/M3MVe89wdSmaYfowZU/KWl2GrWxcOPdT1M3/2GVjrJi6syK5dxXMQKiCUGiO4hVDdxSISo1innRkfp3JIDbZiBVwzbC51Azee6X3+/hx/fAwn9LqNV6yAcGZiX7LEdTGD5iCUmiM4IFRCiYjEKNYxhHcbY2ZWsP0nY0zIewJLeti40cVve6zzE0qGXNsvtpN6ASGEN/2MppyRmmi33eAvf4EmTeCMM5JdGhGp5mIdQzgCqChr9xNcdrGkodxcdxeSX36BfriA0AbuUBKTLl1gzz3dcjjjCDXljNRUY8a4idr33z/ZJRGRai7WgLA98FsF25fg5v6TNFNQAKedBt9/754fuZsLCE3fvpCZGdvJjfFbCSdMgB07Kt7fCwgbN3bz6onUJLH+P4mIEHtAuJ2KA76OuAmqJdV9+y288opL0oiDf/4T/vc/t3zUkF10ynEJJfSLsbvY4wWEeXkuKKyI12XcpUvlCSgiIiJpKNaAcDxwsTGmzEy/xph2uNvOfRPjNSTR1q+Ho46Cc86B99+P+XRffw333uuW990X3rlzLsa7j3C8urYOOwxqBZLkK+s21pQzIiIiFYo1ILwVyALmGmMeMsacH3j8B/gZqBPYR1LZ9OmupQ3gk09iOtUff8CZZ7qGxnr13L2Ksxf4CSVxayFs1Mift6aixJKcHPj9d7esgFBERCSkWKedWQAMwk04fTXwXOBxFTATGGStDWNeEEmq2bP95cq6XytgLZx3Hqxe7Z4/9lhgNgzvlnXxSCgJ5nUbL1gAS5eG3mfRIn9ZU86IiIiEFPM9vKy1s621h+LuU3xQ4NEicEu52RUfLSkhOCD89Vc/oovQY4/Bxx+75VNOgQsuCGzwAsJ4JJQE8+5rDOV3G2vKGRERkUrF7aau1tr11tqpgcf6eJ1XqsDsUnF7FK2EP/0E11/vlvfcE555JpC/kZ/vnz9e3cWePn2gRQu3XF63cfCUM2ohFBERCSnWW9cBYIzZA+gLNCZEkGmtfSUe15EEyM8ve7eP8ePdnDFh2r7d7b5rl2sAfOMNN1cuAHPn+uMT4z1XWkaGayV89VX46iv3WmrXLrmPFxC2bOnGHYqIiEgZMQWExpi6wMvAybhA0ALevB7B85coIExVCxa4QApcgFVUFHEL4eWX+3HXnXfCwQcHbfzxR3853i2E4AeE27bB5MkweHDJ7cowFhERqVSsXcb3ACcBNwNDcMHgOcBRuDuYzAJ6x3gNSaTg7uITT3Q/58+HtWvDOnz0aHjpJbc8dCjccEOpHRKVUOI56ih/bsFQ3cbBcxCKiIhISPG4dd2L1tr7gLmBdSuttV9aa4cDm4G/xXgNSSQvIMzMhEsu8dd/+22lh/72m39Is2bw2mshckYSlVDiad7cb3ksHRBu3OjmWAQFhCIiIhWINSBsAUwNLOcGfjYI2v4OrgVRUtXPP7ufXbu67tasLPd8/PgKD9u1C04/3fXUgmslbNOm1E6JTCgJ5mUb//RTyZbN4AxjJZSIiIiUK9aAcC3QDMBauwPYBHQN2t4IqBvjNSSRvICtVy+oWxcOPNA9r2Qc4a23wrRpbvnKK2H48BA7JTKhJJg3HyHA55/7y5pyRkREJCyxBoRTgIFBzz8E/m6MGWmMOQs3WfUPMV5DEmXTJv8uHr16uZ9Dhrifc+f63a2lfP453H+/W+7TB+67r5zzJzqhxHPQQdC4sVsO7jb2EkqMgU6dEnd9ERGRai7WgPAxYLExJtDPyK24cYOv4rKPtwBXxHgNSRSvuxigZ0/389BD/XUhxhGuXQtnn+2W69eHN9/0e5nLSHRCiadWLTj8cLf8+ecuUxr8gLB9e9f6KSIiIiHFeuu676y1V1pr8wLPVwDdcXMS9gK6B25vJ6koOMPYayE86CB/Lr9S4withQsv9IfpPfGEG3pYrkQnlATzuo3Xr4cZM9yyppwREREJS9QBoTGmvjHmXWPMyOD11toia+0sa+0ca21B7EWUhPECwsaNoV07t1y/PvTv75ZLjSN86y346CO3fOqpcO65FZy7qhJKPMG3sfvsMxe9asoZERGRsEQdEAaSSI4A6sevOFKlghNKvLn8wB9H+PPPbuoW3I8rAp3/u+/uWgeDDymjqhJKPO3bQ/fubnncOFizxt1CBRQQioiIVCLWMYTfAQdXupeknqIimDPHLXvjBz3eOEJrYeJEAK67Dtatc6sffdQFhRX67jt/uSpaCMHvNp482U+BBk05IyIiUolYA8LLgEHGmLsD9zOW6mLJEsjJccve+EHPgAEuUQNg/Hi+/BJefNE9PfpoN/9ghayF5593y+3bVzLQMI68gLCwEEaN8terhVBERKRCsQaEs4A9gBuBZcaYPGPM1lKPLbEXU+IuVEKJp0GD4m7eom8mcPHF/ur//reSrmJw083MnOmWL7zQ3SO5Kgwa5GcTe9PP1K4Ne+5ZNdcXERGppmrFePw7gI1HQaSKBQeE++5bdvuQIfDDDzBrJhvYDDThX/8KM7Z65hn3MzMTzj8/9rKGq149V+7guQg7dfJbO0VERCSkmD4prbXnxqkcUtW8OQg7doSGDctuP/RQ+Pe/ycAykO/4o/9wLrssjPNu3QpvvOGWhw+Htm3jVuSwHH10yYBQ4wdFREQqVUV9eZJygjOMQyg48BAKcHMHHmbG89xzYU4lOHq0PzbxooviUNAIBU8/Axo/KCIiEoaYWgiNMWeHs5+19pVYriNxlpMDixa55XICwkeeb8hg9qM/0/hLqwns0TPkbiVZC08/7Zbbty8bnFWFrl1dv/ayZe65AkIREZFKxTq46qUKtgWPLVRAmErmznXBG5Sdcgb47Te47Ta4nSH0Zxpt185wXcGNGlV83tLJJIm+O0koxrhuYy8wVZexiIhIpWLtMu4Y4rE3bsLq94DpQIiMBUmq4HsYl2ohtBYuvhhyc2ECbj5CU1QE339f+XmTlUxS2imnuJ/160Pv3skrh4iISDUR672Ml4V4LLbWfm2tHQH8gZurUFKJN36wbl3Ye+8Sm155Bb76yi3vfe5Af8qYUrexKyPZySTBDj8cvvgCJk2C3XZLXjlERESqiUQnlXwEnJrga0ikvIBw331LdOuuXQtXX+2WW7eGOx5uDH36uBXjx1d8zmQnk5R2xBFqHRQREQlTogPCTkBWgq8hkbDWDwhLjR+86irYtMktP/kkNGmCf1/jH3/07w0c6pzJTiYRERGRqMWaZTy4nE1NgMHAFcD7sVxD4mz1ati40S0HjR/8+GN48023/Oc/uwfg5iP8z3/c7eAmTYKjjip7zunTk59MIiIiIlGLNct4PKHvVGKAQuBt4PIYryHxFOKWdbt2waWXulWNGsETTwTtP2iQy9y11o0jDBUQeq2DyU4mERERkajEGhAODbHOApuAZdbarTGeX+ItOCAMdBlPnAjLl7tV//oXtGkTtH/Tpi5wnDUr9DjCVEomERERkajEeuu6SlJPJeV4AWGrVtC8OQCff+5WGQNnnBHimCFDXEA4bRrs2OGmc/GkWjKJiIiIRCympBJjTEdjzPEVbD/eGNMhlmtInHlzEAaNH/QCwgMOKGeWlkPdfITk58Pkyf56JZOIiIjUCLFmGT+ISxwpz9+Af8d4DYmXXbtg3jy3HAgI167180HKjecGDfKXg+cjVDKJiIhIjRBrQHgw8EUF278CBlWwXarSggWulQ+Kxw9+EVR7ofJFANh9dzdnIZQcR+i1DmZkKJlERESkGos1IGwKbKtg+3agWYzXkHgJkWHsdRc3bAgHHljBsd58hFOmuPvaKZlERESkxog1IFwOHFLB9kHA7zFeQ+LFGz+YmQndu2OtHxAedhjUrl3Bsd44wl27XFAYnExy8cUJK7KIiIgkXqwB4RvA6caYK4wxxecyxmQaY67E3bZudIzXkHjxWgi7dYOsLH7+2Y0hhDDyQQYHzUE+frySSURERGqQWOchvBcYCDwC3GyMWRBY3xVojpu4+l8xXkPipdQt68aN8zeVO37Q06IFdO/uklKefRZWrXLrlUwiIiJS7cXUQmitzQOOAi4ApgK7Bx5TgfOBIwL7SLJt3AgrV7rlUuMH99oLOnUK4xzeOEIvGFQyiYiISI0Qawsh1toi4MXAQ1KVN34QoFcvduxwdyiBMFoHPYceCqNG+c+VTCIiIlIjxDox9W7GmF4VbO9pjGkayzUkTkplGE+cCHmBttuwhwB6iSUeJZOIiIjUCLEmlTwMPFPB9qdxk1dLsnkBYePGsMcexeMHMzNhaKg7UofSqhV07eqWlUwiIiJSY8QaEB4G/K+C7R8CR8R4DYkHLyDs1QuMKR4/eNBBLkYM2/33Q79+LstYySQiIiI1QqxjCJsD6yvYvgFoEeM1JFZFRTBnjlvu1YuVK2HuXPc07PGDnhNOcA8RERGpMWJtIVwN9K1gez/gjxivIbFavBh27HDLPXuWuF2den1FREQk1oDwfeACY0yZJiNjzInAecB7MV5DYlUqocQbP9ikCey/f1JKJCIiIikk1i7j23FjBN8zxswCAv2S7Av0AX4B/hnjNSRWQQFh0T77FrcQHnGEhgGKiIhI7BNTbwEOAu4GagMjAo/awJ1Af8DEWEaJlTcH4V578dOihmzY4J6qu1hEREQg9i5jrLU51tp/Wmt7WmvrW2vrAwcAc3H3MV4d6zUkRkG3rPOyiwGOPDI5xREREZHUEvOdSjzGGAMcDowE/gw0xGUgj47XNSQK27fDb7+55aDxg127wp57Jq9YIiIikjpibiE0xvQzxvwHWAl8DpwNfAwMBFpZa6O+2a0xJssYc58xZpUxJtcYM8UYE3G7ljHmC2OMNcY8EW1Zqq25c8FaAHI792LSJLc64ulmREREpMaKKiA0xuxljLnVGDMfmIobN/g6cCpuzOA71trJ1gYikei9BFwTOPeVQCHwiTFmYARlPQk4OMZyVF9B9zD+YUcv8vPdssYPioiIiCfiLmNjzGRcssh6YCxwobX2u8C2TvEqmDGmP3Aa8Hdr7YOBda/gMpnvBwaEcY66wEPAfbgkl/TjjR+sV4/3f3bVU7t22dsSi4iISPqKpoXwQGApcBFwpRcMJsAIXItg8b2SrbU7geeBg40x7cI4x/W415i+91P2AsIePfjsCzfHzCGHQHZ2EsskIiIiKSWagPAyXObwe8AaY8zTxpihgaSSeOoLLLTWbi21fmrgZ5+KDjbGtAf+Adxgrc2Nc9mqB2uLA8Jte/Vi4UK3WuMHRUREJFjEXcbW2qeAp4wxHXEZxWcAfwXWAN8ANvCIVWtCT1njrWtTyfEPAT9Za9+M9MLGmBa4+zQH6wSwfft2tm4tHaPGR05OTomfsTKrVtFw0yYAZuR3L15/yCHb2bq1KC7XSCfxrh+JH9VNalP9pC7VTWqLpX62b98e0f5RTztjrV2Cm5D6bmNMP1xw6CWVPGWMOQb4H/BloKs3UvWAvBDrdwZtD8kYMxQ4Gde9HY1LKecOK1OnTmXNmjVRnjY8U6dOrXynMLSYPr04m2bswvYANGqUx4YNX/HNN3G5RFqKV/1I/KluUpvqJ3WpblJbNPWzfPnyiPaPyzyE1trpwHRjzHXAYcCZuODwQmAHEM2ItVwgK8T6ukHbyzDG1AIeA1611k6L4roATwFvl1rXCfigf//+dO/ePcQhscvJyWHq1Kn079+fBg0axHy+OjNnFi9/snwIAEcdlcHhhw+N+dzpKN71I/Gjukltqp/UpbpJbbHUz7x58yLaP24TUwNYa4uAL4EvjTGXACfiupSjsRpoG2J968DPVeUcdzbQFbjYGNOh1LaGgXXrrLU7yruwtXYdsC54nTdEMjs7m0aNGlVW9pg0aNAgPteYPx+AXbu3YfH6FgAMH16bRo1qx37uNBa3+pG4U92kNtVP6lLdpLZo6ic7wuzRmCemLo+1dqe1doy19sQoTzET6GKMKf0bODBoeyjtcfdS/h5YEvQAFywuAdIjrWLWLACWNe1TvEq3qxMREZHS4tpCGGdjgetw09t48xBmAecBU6y1KwLr2gP1rbXzA8e9Sehg8T3gE+BZYEpCS54KduyABQsAmJzbB4B994U2laXiiIiISNpJ2YDQWjvFGPM2cG8g63cRcA7QAbggaNdXgENxySwEAsP5lBLo8l1irX0/oQVPFXPmQJHLJP5kZW9A082IiIhIaCkbEAacDdwFnAU0BWYDw6213ya1VNVBUELJDNsH0O3qREREJLSUDggD09X8PfAob58hYZ4r3hNnp7bA+MGdtRrwW0EnsrJg0KAkl0lERERSUsKSSiTJAi2Ev2T2oohMBg+GeuXO3CgiIiLpTAFhTVRUVNxCOCVP4wdFRESkYgoIa6LffoPAbW5mBm75rPGDIiIiUh4FhDVRUELJTPqwxx5uyhkRERGRUBQQ1kSB7uJCMviZnpxyCpj0SqkRERGRCCggrIkCLYQL6UIu9fnLX5JbHBEREUltCghrokBAOIvetGsHBx5Y8e4iIiKS3hQQ1jTr18PKlYAbP6juYhEREamMAsKaJjB+EPyAUERERKQiCghrmqAM4/Vt+6i7WERERCqlgLCG2TF5JgBraMmhp7ZSd7GIiIhUSgFhDZM7aSbgEkrUXSwiIiLhUEBYk+zcSePV8wBY3FDdxSIiIhIeBYQ1yB8TfqEWhQA0HNRH3cUiIiISFgWENcjsV2YWL/c8q0/SyiEiIiLViwLCGmTz+JkA7DR16XVy5+QWRkRERKoNBYQ1xOrV0HzVTADWteyJqV0ruQUSERGRakMBYQ3xzlhLb9yk1HUP7JPcwoiIiEi1ooCwhpj46lIasxWA5kf2SW5hREREpFpRQFgDrFoF+dNmFj83ffskrSwiIiJS/SggrAHeeQd6M9Nf0bNn0soiIiIi1Y8CwhrgrbcoHj9o994bGjZMcolERESkOlFAWM2tXAnffw99Ai2Epk+fpJZHREREqh8FhNXcO+9AY7uJDixzKxQQioiISIQUEFZzb7/tdxcDCghFREQkYgoIq7GVK+G77/zuYgB6905aeURERKR6UkBYjb3zjvtZ3ELYrBm0bZu8AomIiEi1pICwGnvrLfezf52ZbqFPHzAmWcURERGRakoBYTXlZRfXZhddC+a6lRo/KCIiIlFQQFhNjR3rfnZnHrWK8t0TjR8UERGRKCggrKbeftv9PGJ3ZRiLiIhIbBQQVkO//+66iwFO3HOmW6hTB7p1S1qZREREpPpSQFgNednFAH3MTLew775Qu3ZSyiMiIiLVmwLCasjrLu6wp6XhbzPdE3UXi4iISJQUEFYzGzbA5Mlu+fwjV2A2bXJPlFAiIiIiUVJAWM189hkUFbnlEzsooURERERip4Cwmvn4Y/ezUSPokT/T36AWQhEREYmSAsJqpKAAPv3ULR91FGT+PNM96dgRGjdOWrlERESkelNAWI1MngybN7vl4cOBmTPdE3UXi4iISAwUEFYjXnexMXDswK2weLFboe5iERERiYECwmrECwgPOACar57tb1ALoYiIiMRAAWE1sXw5zJnjlo87Dr+7GBQQioiISEwUEFYTXusglAoImzSB9u2TUCIRERGpKRQQVhNeQNiqFfTtC8wKzEHYu7cbVCgiIiISJQWE1cCOHfDVV2752GMho6gAfv7ZrVB3sYiIiMRIAWE18M03sHOnWz7uOGDBAsjLcysUEIqIiEiMFBBWA153ce3acOSRKKFERERE4qpWsgsgAfn5mNWrqbtxI2b1ati+HQBrYdoH0BoYfBA03AZMmuSOqV0b9tknaUUWERGRmkEBYaqYN4+GvXszrNRqA0zznkwE2gZt7N4d6tSpitKJiIhIDaYu4+rs8MOTXQIRERGpAdRCmCratiX3kUdYsGABXbt2pV69egA88AD8ughatoC77grav3FjOP745JRVREREahQFhKmiWTPyzzuPZd98w15Dh1KvUSM2bYIbL4VC4KozgIuSXUgRERGpidRlnMLGjYPCQrd83HHJLYuIiIjUXAoIU9hHH7mf2dkweHByyyIiIiI1lwLCFFVYCJ995paPOkrJxCIiIpI4CghT1JQpsGGDW1Z3sYiIiCSSAsIU5d2dBNz9i0VEREQSRQFhivICwn79oFWr5JZFREREajYFhClo5UrDrFluWd3FIiIikmgpHRAaY7KMMfcZY1YZY3KNMVOMMUeGcdxJxpgxxpjFxpgdxpgFxpiHjDFNqqDYMfv8c396SAWEIiIikmgpHRACLwHXAK8DV+LmaP7EGDOwkuOeAboDrwFXAJ8BlwGTjTH1ElbaOPECwhYtYP/9k1wYERERqfFS9k4lxpj+wGnA3621DwbWvQLMAe4HBlRw+Ahr7fhS55sOvAyMBJ5LRJnjYdeuDMaPd9Vy7LGQkeohu4iIiFR7qRxujMC1CD7jrbDW7gSeBw42xrQr78DSwWDAe4Gf3eNYxribM2d3duwwgLqLRUREpGqkbAsh0BdYaK3dWmr91MDPPsCKCM7n5equr2xHY0wLoHmp1Z0Atm/fztatpYsUHzk5Ofz4Y0sAatWyHHjgNhJ0KYlCTk5OiZ+SOlQ3qU31k7pUN6ktlvrZvn17RPunckDYGlgdYr23rk2E57sB1+I4Nox9LwX+GWrD1KlTWbNmTYSXDo+1MH36EQB0776eGTMmJeQ6EpupU6dWvpMkheomtal+UpfqJrVFUz/Lly+PaP9UDgjrAXkh1u8M2h4WY8wZwAXA/dbaX8M45Cng7VLrOgEf9O/fn+7dE9PrPHNmHmvXNgDg1FMbMnTo0IRcR6KTk5PD1KlT6d+/Pw0aNEh2cSSI6ia1qX5Sl+omtcVSP/PmzYto/1QOCHOBrBDr6wZtr5QxZhBu3OE44OZwjrHWrgPWlToPANnZ2TRq1Cic00Ts2293Fi+ffHJdGjWqW8HekiwNGjRI2N+AxEZ1k9pUP6lLdZPaoqmf7OzsiPZP5aSS1bhu49K8dasqO4ExpjfwP1xm8ghrbUH8ihd/3nQzHTsW0bVrkgsjIiIiaSOVA8KZQBdjTOmQ+MCg7eUyxnTCzT+4DjjWWhvZ6MoqtmULTJ6cCcCwYfkEGiRFREREEi6VA8KxQCZwkbfCGJMFnAdMsdauCKxrb4zpFnygMaYV8DlQBAyz1v5RZaWO0uefQ0GBiwKPOiqlGzJFRESkhknZMYTW2inGmLeBewPTwCwCzgE64BJEPK8AhwLBbWqfAXvhJrAeWOrOJmuttV8ksuzROPZYePPNHbz88loGDmyW7OKIiIhIGknZgDDgbOAu4CygKTAbGG6t/baS43oHfl4fYtsEIOUCwgYN4JhjCqhbdzZZWcouFhERkaqT0gFh4M4kfw88yttnSIh1GoEnIiIiEqZUHkMoIiIiIlVAAaGIiIhImlNAKCIiIpLmFBCKiIiIpDkFhCIiIiJpTgGhiIiISJpTQCgiIiKS5hQQioiIiKQ5BYQiIiIiaU4BoYiIiEiaU0AoIiIikuYUEIqIiIikOQWEIiIiImlOAaGIiIhImlNAKCIiIpLmFBCKiIiIpDkFhCIiIiJpTgGhiIiISJpTQCgiIiKS5hQQioiIiKQ5BYQiIiIiaU4BoYiIiEiaU0AoIiIikuYUEIqIiIikOQWEIiIiImlOAaGIiIhImlNAKCIiIpLmFBCKiIiIpDkFhCIiIiJpTgGhiIiISJpTQCgiIiKS5hQQioiIiKQ5BYQiIiIiaU4BoYiIiEiaU0AoIiIikuYUEIqIiIikOQWEIiIiImlOAaGIiIhImlNAKCIiIpLmFBCKiIiIpDkFhCIiIiJpTgGhiIiISJpTQCgiIiKS5hQQioiIiKQ5BYQiIiIiaU4BoYiIiEiaU0AoIiIikuYUEIqIiIikOQWEIiIiImlOAaGIiIhImlNAKCIiIpLmFBCKiIiIpDkFhCIiIiJpTgGhiIiISJpTQCgiIiKS5hQQioiIiKQ5BYQiIiIiaU4BoYiIiEiaS+mA0BiTZYy5zxizyhiTa4yZYow5Msxj2xpj3jLGbDbGbDXGfGCM2SvRZRYRERGpblI6IAReAq4BXgeuBAqBT4wxAys6yBiTDXwDHArcA/wT6AtMMMY0S2SBRURERKqbWskuQHmMMf2B04C/W2sfDKx7BZgD3A8MqODwS4HOQH9r7bTAsZ8Gjr0WuCmBRRcRERGpVlK5hXAErkXwGW+FtXYn8DxwsDGmXSXHTvOCwcCx84GvgL8kprgiIiIi1VMqB4R9gYXW2q2l1k8N/OwT6iBjTAbQC/gxxOapQCdjTMN4FVJERESkukvZLmOgNbA6xHpvXZtyjtsNyArj2AXlXdgY0wJoXmp1N4DZs2ezffv28g6NSW5uLsuXL2fGjBnUq1cvIdeQ6Kl+UpfqJrWpflKX6ia1xVI/ixcv9hbrhLN/KgeE9YC8EOt3Bm0v7ziiPNZzKS4RpYwzzjijkkNFREREUkY74KfKdkrlgDAX19JXWt2g7eUdR5THep4C3i61LhvogktM2VXJ8dHqBHwAnAj8lqBrSPRUP6lLdZPaVD+pS3WT2mKpnzq4YHBCODunckC4GmgbYn3rwM9V5Ry3Edc62DrEtsqOBcBauw5YF2LTlIqOi5Uxxlv8zVo7N5HXksipflKX6ia1qX5Sl+omtcWhfiptGfSkclLJTKCLMaZRqfUHBm0vw1pbBPwM7B9i84HAYmvttjiVUURERKTaS+WAcCyQCVzkrTDGZAHnAVOstSsC69obY7qFOPYAY8z+Qcd2BQ6jbFewiIiISFpL2S5ja+0UY8zbwL2BrN9FwDlAB+CCoF1fwd2RxAStewr4K/CxMeZBIB93x5O1wEOJL72IiIhI9ZGyAWHA2cBdwFlAU2A2MNxa+21FB1lrtxljhgAPA7fgWkLHA1dba/9IYHlj9QdwR+CnpB7VT+pS3aQ21U/qUt2ktiqrH2OtTfQ1RERERCSFpfIYQhERERGpAgoIRURERNKcAkIRERGRNKeAUERERCTNKSAUERERSXMKCFOAMSbLGHOfMWaVMSbXGDPFGHNkssuVbowx2caYO4wxnxljNhpjrDHm3HL27R7Yb3tg31eNMc2ruMhpwxhzgDHmCWPMXGNMjjFmuTHmLWNMlxD7qm6qkDGmhzHmbWPMYmPMDmPMemPMt8aY40Psq7pJMmPMzYH3tjkhtg0wxnwXqMc1xpjHjDHZyShnOjDGDAnURajHQaX2TXjdpPo8hOniJWAE8AjwK3Au8IkxZqi19rvkFSvt7A7cBiwHZgFDQu1kjNkD+BbYAtwEZAPXAT2NMf2ttbuqpLTp5QbgENydhmYDrYDLgBnGmIOstXNAdZMkewINgZdx94mvD5wM/M8Yc7G19hlQ3aSCQB3cBOSE2NYH+AqYh7uRwx64+ukMHFN1pUxLjwHTSq1b5C1UWd1Ya/VI4gPoD1jguqB1dQN/DJOSXb50egBZQKvA8v6Bejk3xH5PATuA9kHrjgjsf1GyX0dNfAADgDql1nUGdgKvqW5S64G77ehMYL7qJnUewJu4wGI8MKfUtk9wAX2joHUXBurnqGSXvSY+cI0OFhhRyX5VUjfqMk6+EUAh8Iy3wlq7E3geONgY0y5ZBUs31to8a+2aMHY9GfjIWrs86NgvgYXAXxJVvnRmrZ1kS7UgWWt/BeYC3YNWq25SgLW2EFgBNAlarbpJImPMYNznzVUhtjUCjsR9udoatOkVYDuqn4QzxjQ0xpTpta3KulFAmHx9gYWlKhpgauBnn6otjlTEGNMWaAH8GGLzVFx9ShUwxhigJbA+8Fx1k0TGmAbGmN2NMZ2MMVfjurK+CmxT3SSRMSYTeBx4zlr7c4hdeuKGkJWon8CXsJmofhLtRWArsNMY840xZv+gbVVWNxpDmHytgdUh1nvr2lRhWaRyrQM/y6uz3YwxWdbavCosU7oaCbTFjfsE1U2yPQRcHFguAt7FjfME1U2yXYIb63lEOdsrq59BiSiUsAt4B9clvB7YBzc2cKIxZoC19ieqsG4UECZfPSDUm+DOoO2SOrz6qKzO9MGWQMaYbsCTwGRcMgOobpLtEWAs7kvsX3DjCOsEtqluksQY0wy4E7jLWvtHObtVVj/6HEoAa+0kYFLQqv8ZY8biEufuBY6mCutGXcbJl4tLZiitbtB2SR1efajOksQY0wr4GJetOiIwXg1UN0llrZ1vrf3SWvuKtXY4Lov4w0DXvuomee4GNuK6jMtTWf2obqqItXYR8AEwNNDVX2V1oxbC5FuN6/YqzWsmXlWFZZHKec32rUNsaw1sVLdX4hhjGgOf4pIVBllrg/8/VDepZSzwNNAF1U1SGGM6AxfhEknauNgccIFEbWNMB9zYtcrqR59DVWsFrnW9AVVYN2ohTL6ZQJdAJlGwA4O2S4qw1q4E/sBNS1Naf1RfCWOMqQt8iAswhltrfwnerrpJOV5XVmPVTdK0xX3OPwYsCXociPs/WoIbgzsHKKBU/Rhj6uASG2dWVYEFgL1w3cHbqcK6UUCYfGNxY20u8lYYY7KA84Ap1toVySqYlOsdYHjwlEDGmMNxb7BvJ61UNVig62QMcDBwirV2cjm7qm6qmDGmRYh1tYGzcd1ZXuCuuql6c4A/h3jMxU3A/2fgeWvtFuBL4ExjTMOg48/Cdf2rfhIg1F16jDG9gROAz621RVVZNyYwwaEkkTHmLdw/5sO4CanPwX1rPtxa+20yy5ZujDGX4boj2wD/h8uU/Cmw+XFr7ZbAB9pPwGbgUdw/5d+B34ED1PUVf8aYR4ArcS2Eb5Xebq19LbCf6qaKGWPeAxrh7kKyEncXmZFAN+Baa+1/AvupblKEMWY8sLu1dt+gdfvhEhx+wc2LuwdwLfCttXZYMspZ0xljvsZ9aZoErMNlGV8E5AMHW2vnBfarkrpRQJgCAl1hdwFnAk1xGUa3WmvHJbVgacgYsxQ3PUMoHa21SwP79QD+AwzETR3wMe7Db20VFDPtBD7ADi1vu7XWBO2ruqlCxpjTgAtw86U1A7YB03FfoP5Xal/VTQoIFRAG1g8E7gP2w9XjW8CN1tptVV7INGCMuQL35Wlv3JeqP3Bzd94RSC4J3jfhdaOAUERERCTNaQyhiIiISJpTQCgiIiKS5hQQioiIiKQ5BYQiIiIiaU4BoYiIiEiaU0AoIiIikuYUEIqIiIikOQWEIiIiImlOAaGIiIhImlNAKCIiIpLmFBCKiFQjxphzjTHWGLN/sssiIjWHAkIRkVKCgq7yHgclu4wiIvFUK9kFEBFJYbcBS0KsX1TVBRERSSQFhCIi5fvUWvtjsgshIpJo6jIWEYmCMaZDoPv4OmPM1caYZcaYXGPMBGPMviH2P8wYM9EYk2OM2WyM+cAY0z3Efm2NMc8bY1YZY/KMMUuMMaOMMXVK7ZpljPmPMeaPwDnfM8Y0L3Wu/Y0x44wx6wNlW2KMeSHOvwoRqQHUQigiUr7GxpjdS62z1toNQc/PBhoCTwJ1gSuBr40xPa21awGMMUcAnwKLgduBesDlwPfGmP2stUsD+7UBpgJNgGeA+UBbYARQH9gVdN3HgU3AHUAH4CrgCeDUwLlaAJ8DfwD/BjYH9jspyt+FiNRgCghFRMr3ZYh1ebjAz7M30NlauxLAGPMZMAW4AbgmsM8DwEbgYGvtxsB+7wM/4QK6cwL73Qu0Ag4s1VV9mzHGlCrHBuAoa60NnC8DuMIY09hauwUYADQN7BN8rlvCfO0ikkbUZSwiUr6/AUeWehxTap/3vWAQwFo7FRcQHgtgjGkN9AFe8oLBwH6zgS+C9ssA/gR8GGrcohf4BXmm1LqJQCawZ+D55sDP4caY2mG9WhFJW2ohFBEp39Qwkkp+DbFuIfCXwLIXoC0Isd88YJgxpgGQDTQC5oRZtuWlnm8K/Gwa+DkBeAf4J3C1MWY88D4w2lqbF+Y1RCRNqIVQRKR6KixnvQHXomitHQEcjBtb2BZ4AZhujMmumiKKSHWhgFBEJDadQ6zrAiwNLC8L/OwaYr9uwHprbQ4u+WMrUCZDORbW2h+stTdba/cHRgI9gNPieQ0Rqf4UEIqIxOZPxpi23hNjTH/gQFxWMdba1cBM4BxjTJOg/fYFjgI+CexXhOvSPT7UbelCJJVUyBjTNMQxMwM/syI5l4jUfBpDKCJSvmOMMd1CrJ8EFAWWFwHfGWNG4QKtq3AZwPcH7f93XIA42RjzPP60M1tw09B4bsIFiROMMc/gxhi2Bk4BBuInioTjHOBSY8x7wG+4qXH+imuF/CSC84hIGlBAKCJSvjvLWX8eMD6w/AouOLwKaIGbR/CyQMsgANbaL40xR+OmmLkTyMclfdxgrV0StN9KY8yBwF247t1GwEpcMLkjwrJPAPrjuodb4oLPqcDI4GuKiACYsjMZiIhIZYwxHXD3Of67tfbBJBdHRCQmGkMoIiIikuYUEIqIiIikOQWEIiIiImlOYwhFRERE0pxaCEVERETSnAJCERERkTSngFBEREQkzSkgFBEREUlzCghFRERE0pwCQhEREZE0p4BQREREJM0pIBQRERFJcwoIRURERNKcAkIRERGRNPf/OR6vCl6sYToAAAAASUVORK5CYII=\n", "image/svg+xml": [ "\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2021-05-09T23:59:37.477771\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.4.1, https://matplotlib.org/\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", " \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", " \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", " \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", " \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" ], "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "history.history\n", "plt.plot(range(len(history.history[\"accuracy\"])), history.history[\"accuracy\"], label=\"Train\", c=(0, 0, 1))\n", "plt.plot(range(len(history.history[\"val_accuracy\"])), history.history[\"val_accuracy\"], label=\"Validation\", c=(1, 0, 0))\n", "\n", "plt.xlabel(\"Epochs\")\n", "plt.ylabel(\"Accuracy\")\n", "plt.ylim(0, 1)\n", "\n", "plt.grid()\n", "plt.legend()\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Test\n", "\n", "Single number below from the evaluate function" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "10/10 [==============================] - 0s 2ms/step - loss: 0.7529 - accuracy: 0.7609\n" ] }, { "data": { "text/plain": [ "[0.7529351115226746, 0.7609427571296692]" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.evaluate(data_test.to_numpy(), labels_test)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Get raw predictions from test data to generate a confusion matrix" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "image/svg+xml": [ "\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2021-05-09T23:59:37.976238\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.4.1, https://matplotlib.org/\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", " \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", " \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", " \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", " \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", " \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", " \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", " \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", " \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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "predictions = model(data_test.to_numpy())\n", "\n", "conf = tf.math.confusion_matrix([tf.math.argmax(i) for i in labels_test], \n", " [tf.math.argmax(i) for i in predictions], \n", " num_classes=len(filtered_playlists))\n", "\n", "normalised_conf = np.ndarray((len(filtered_playlists), len(filtered_playlists)))\n", "for idx, row in enumerate(conf):\n", " normalised_conf[idx, :] = row / np.sum(row)\n", "\n", "sns.heatmap(normalised_conf, \n", " annot=True, \n", " xticklabels=playlist_names, yticklabels=playlist_names, \n", " cmap='inferno')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Ensemble Model" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [], "source": [ "models = [get_model(hidden_nodes=random.randint(16, 128), \n", " layers=random.randint(1, 2)) \n", " for _ in range(9)]\n", "\n", "for m in models:\n", " m.compile(\n", " optimizer=tf.keras.optimizers.Adam(learning_rate=0.01), \n", "# optimizer=tf.keras.optimizers.SGD(learning_rate=0.01, momentum=0.9),\n", " loss='categorical_crossentropy', \n", " metrics=['accuracy'])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Train the models of the meta-classifier. Get a random number of epochs from a reasonable range to introduce variation between models. *Weird?*: Randomly decide whether to regularise by class weights. Class weights change the penalisation methods such that smaller classes are treated more importantly when computing loss. I'm not sure whether it's going to help. " ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "training model 1\n", "training model 2\n", "training model 3\n", "training model 4\n", "training model 5\n", "training model 6\n", "training model 7\n", "training model 8\n", "training model 9\n" ] } ], "source": [ "if BALANCED_WEIGHTS:\n", " cw = class_weights\n", "else:\n", " cw = None\n", "\n", "ensem_histories = list()\n", "for idx, m in enumerate(models):\n", " print(f'training model {idx+1}')\n", " h = m.fit(data_train.to_numpy(), labels_train, \n", " callbacks=[tensorboard_callback()], \n", " validation_split=0.11,\n", " verbose=0,\n", "# class_weight=cw,\n", " class_weight=random.choice([*([class_weights]*3), None]),\n", " epochs=random.randint(20, 100))\n", " ensem_histories.append(h)" ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "79.5% Accuracy, 82.7% Agreement, 67.6% Ind. Accuracy\n" ] } ], "source": [ "ensem_results = ensem_classify(models, data_test, labels_test)\n", "print(f\"{ensem_results[2]*100:.3}% Accuracy, {ensem_results[3]*100:.3}% Agreement, {ensem_results[4]*100:.3}% Ind. Accuracy\")" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "image/svg+xml": [ "\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2021-05-10T00:00:22.760673\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.4.1, https://matplotlib.org/\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", " \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", " \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", " \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", " \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", " \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", " \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", " \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", " \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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "ensem_conf = tf.math.confusion_matrix([tf.math.argmax(i) for i in labels_test], \n", " ensem_results[0], \n", " num_classes=len(filtered_playlists))\n", "\n", "normalised_ensem_conf = np.ndarray((len(filtered_playlists), len(filtered_playlists)))\n", "for idx, row in enumerate(ensem_conf):\n", " normalised_ensem_conf[idx, :] = row / np.sum(row)\n", "\n", "sns.heatmap(normalised_ensem_conf, \n", " annot=True, \n", " xticklabels=playlist_names, yticklabels=playlist_names, \n", " cmap='inferno')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Imports & Setup" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "from datetime import datetime\n", "import os\n", "import random\n", "\n", "from google.cloud import bigquery\n", "import matplotlib.pyplot as plt\n", "import matplotlib as mpl\n", "mpl.rcParams['figure.dpi'] = 120\n", "import seaborn as sns\n", "\n", "from analysis.nn import ensem_classify\n", "from analysis.net import get_spotnet, get_playlist, track_frame\n", "from analysis.query import *\n", "from analysis import spotify_descriptor_headers, float_headers, days_since\n", "\n", "import numpy as np\n", "import pandas as pd\n", "\n", "from sklearn import svm\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.metrics import plot_confusion_matrix\n", "from sklearn.utils import class_weight\n", "\n", "import tensorflow as tf\n", "\n", "client = bigquery.Client()\n", "spotnet = get_spotnet()\n", "cache = 'query.csv'\n", "first_day = datetime(year=2017, month=11, day=3)\n", "sig_max, c_max = 0.5, 20" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Read Scrobble Frame" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "scrobbles = get_query(cache=cache)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Write Scrobble Frame" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "scrobbles.reset_index().to_csv(cache, sep='\\t')" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.9" }, "metadata": { "interpreter": { "hash": "bce1a3677099e73bf385a0de8ef462673e03f7df0abce93e57e7ca76e8c504a2" } } }, "nbformat": 4, "nbformat_minor": 4 }