{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"Deep Learning con Keras - Lezione 3 - parte 1","provenance":[],"collapsed_sections":[],"authorship_tag":"ABX9TyMPEHmEB+QByCMjuaTrPtP0"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"code","execution_count":2,"metadata":{"id":"i52hvbIsnkTO","executionInfo":{"status":"ok","timestamp":1647962750952,"user_tz":-60,"elapsed":5826,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}}},"outputs":[],"source":["import numpy as np\n","import tensorflow as tf\n","import matplotlib.pyplot as plt"]},{"cell_type":"markdown","source":["# Classificazione delle recensioni di film su IMDB\n","\n","Lavoreremo col dataset IMDB. Si tratta di 50.000 recensioni di film molto polarizzate, tradde dall'Internet Movie Database. Sono divise in 25.000 recensioni per l'addestramento e altre 25.000 per il test."],"metadata":{"id":"XuyEmJWk5hfw"}},{"cell_type":"code","source":["# Caricamento dati.\n","# Il parametro num_words=10000 vuol dire che teniamo traccia solo delle 10.000\n","# parole più comuni nelle recensioni. Tutte le altre parole vengono ignorate.\n","\n","imdb = tf.keras.datasets.imdb\n","(imdb_train_data, imdb_train_labels), (imdb_test_data, imdb_test_labels) = imdb.load_data(num_words=10000)"],"metadata":{"id":"QL-JsLLe5uN4","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1647962759521,"user_tz":-60,"elapsed":6502,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"4ba85cbc-a494-42af-b7fa-d0f4b109c13e"},"execution_count":3,"outputs":[{"output_type":"stream","name":"stdout","text":["Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/imdb.npz\n","17465344/17464789 [==============================] - 0s 0us/step\n","17473536/17464789 [==============================] - 0s 0us/step\n"]}]},{"cell_type":"markdown","source":["## Preparazione dei dati"],"metadata":{"id":"5U4UVU8n_zSa"}},{"cell_type":"code","source":["# imdb_train_labels è un elenco di 0 ed 1, dove 0 sta per recensione positiva, 1 per negativa\n","\n","imdb_train_labels"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"KSdc4CdT7VSP","executionInfo":{"status":"ok","timestamp":1647962793217,"user_tz":-60,"elapsed":281,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"cc8e237e-b1d8-4853-be72-a9513c320740"},"execution_count":4,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([1, 0, 0, ..., 0, 1, 0])"]},"metadata":{},"execution_count":4}]},{"cell_type":"code","source":["# imdb_train_data è un elenco di recensioni. Ogni recensione è codifica come un elenco di numeri. I numeri\n","# da 0 a 2 hanno un significato speciale, quelli da 3 in poi corrispondono ognuno ad una parola diversa.\n","\n","imdb_train_data"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Om7NB0CR6uE6","executionInfo":{"status":"ok","timestamp":1647962794557,"user_tz":-60,"elapsed":5,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"bbf3e408-a214-4075-d7d9-c8708ebdc14d"},"execution_count":5,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([list([1, 14, 22, 16, 43, 530, 973, 1622, 1385, 65, 458, 4468, 66, 3941, 4, 173, 36, 256, 5, 25, 100, 43, 838, 112, 50, 670, 2, 9, 35, 480, 284, 5, 150, 4, 172, 112, 167, 2, 336, 385, 39, 4, 172, 4536, 1111, 17, 546, 38, 13, 447, 4, 192, 50, 16, 6, 147, 2025, 19, 14, 22, 4, 1920, 4613, 469, 4, 22, 71, 87, 12, 16, 43, 530, 38, 76, 15, 13, 1247, 4, 22, 17, 515, 17, 12, 16, 626, 18, 2, 5, 62, 386, 12, 8, 316, 8, 106, 5, 4, 2223, 5244, 16, 480, 66, 3785, 33, 4, 130, 12, 16, 38, 619, 5, 25, 124, 51, 36, 135, 48, 25, 1415, 33, 6, 22, 12, 215, 28, 77, 52, 5, 14, 407, 16, 82, 2, 8, 4, 107, 117, 5952, 15, 256, 4, 2, 7, 3766, 5, 723, 36, 71, 43, 530, 476, 26, 400, 317, 46, 7, 4, 2, 1029, 13, 104, 88, 4, 381, 15, 297, 98, 32, 2071, 56, 26, 141, 6, 194, 7486, 18, 4, 226, 22, 21, 134, 476, 26, 480, 5, 144, 30, 5535, 18, 51, 36, 28, 224, 92, 25, 104, 4, 226, 65, 16, 38, 1334, 88, 12, 16, 283, 5, 16, 4472, 113, 103, 32, 15, 16, 5345, 19, 178, 32]),\n","       list([1, 194, 1153, 194, 8255, 78, 228, 5, 6, 1463, 4369, 5012, 134, 26, 4, 715, 8, 118, 1634, 14, 394, 20, 13, 119, 954, 189, 102, 5, 207, 110, 3103, 21, 14, 69, 188, 8, 30, 23, 7, 4, 249, 126, 93, 4, 114, 9, 2300, 1523, 5, 647, 4, 116, 9, 35, 8163, 4, 229, 9, 340, 1322, 4, 118, 9, 4, 130, 4901, 19, 4, 1002, 5, 89, 29, 952, 46, 37, 4, 455, 9, 45, 43, 38, 1543, 1905, 398, 4, 1649, 26, 6853, 5, 163, 11, 3215, 2, 4, 1153, 9, 194, 775, 7, 8255, 2, 349, 2637, 148, 605, 2, 8003, 15, 123, 125, 68, 2, 6853, 15, 349, 165, 4362, 98, 5, 4, 228, 9, 43, 2, 1157, 15, 299, 120, 5, 120, 174, 11, 220, 175, 136, 50, 9, 4373, 228, 8255, 5, 2, 656, 245, 2350, 5, 4, 9837, 131, 152, 491, 18, 2, 32, 7464, 1212, 14, 9, 6, 371, 78, 22, 625, 64, 1382, 9, 8, 168, 145, 23, 4, 1690, 15, 16, 4, 1355, 5, 28, 6, 52, 154, 462, 33, 89, 78, 285, 16, 145, 95]),\n","       list([1, 14, 47, 8, 30, 31, 7, 4, 249, 108, 7, 4, 5974, 54, 61, 369, 13, 71, 149, 14, 22, 112, 4, 2401, 311, 12, 16, 3711, 33, 75, 43, 1829, 296, 4, 86, 320, 35, 534, 19, 263, 4821, 1301, 4, 1873, 33, 89, 78, 12, 66, 16, 4, 360, 7, 4, 58, 316, 334, 11, 4, 1716, 43, 645, 662, 8, 257, 85, 1200, 42, 1228, 2578, 83, 68, 3912, 15, 36, 165, 1539, 278, 36, 69, 2, 780, 8, 106, 14, 6905, 1338, 18, 6, 22, 12, 215, 28, 610, 40, 6, 87, 326, 23, 2300, 21, 23, 22, 12, 272, 40, 57, 31, 11, 4, 22, 47, 6, 2307, 51, 9, 170, 23, 595, 116, 595, 1352, 13, 191, 79, 638, 89, 2, 14, 9, 8, 106, 607, 624, 35, 534, 6, 227, 7, 129, 113]),\n","       ...,\n","       list([1, 11, 6, 230, 245, 6401, 9, 6, 1225, 446, 2, 45, 2174, 84, 8322, 4007, 21, 4, 912, 84, 2, 325, 725, 134, 2, 1715, 84, 5, 36, 28, 57, 1099, 21, 8, 140, 8, 703, 5, 2, 84, 56, 18, 1644, 14, 9, 31, 7, 4, 9406, 1209, 2295, 2, 1008, 18, 6, 20, 207, 110, 563, 12, 8, 2901, 2, 8, 97, 6, 20, 53, 4767, 74, 4, 460, 364, 1273, 29, 270, 11, 960, 108, 45, 40, 29, 2961, 395, 11, 6, 4065, 500, 7, 2, 89, 364, 70, 29, 140, 4, 64, 4780, 11, 4, 2678, 26, 178, 4, 529, 443, 2, 5, 27, 710, 117, 2, 8123, 165, 47, 84, 37, 131, 818, 14, 595, 10, 10, 61, 1242, 1209, 10, 10, 288, 2260, 1702, 34, 2901, 2, 4, 65, 496, 4, 231, 7, 790, 5, 6, 320, 234, 2766, 234, 1119, 1574, 7, 496, 4, 139, 929, 2901, 2, 7750, 5, 4241, 18, 4, 8497, 2, 250, 11, 1818, 7561, 4, 4217, 5408, 747, 1115, 372, 1890, 1006, 541, 9303, 7, 4, 59, 2, 4, 3586, 2]),\n","       list([1, 1446, 7079, 69, 72, 3305, 13, 610, 930, 8, 12, 582, 23, 5, 16, 484, 685, 54, 349, 11, 4120, 2959, 45, 58, 1466, 13, 197, 12, 16, 43, 23, 2, 5, 62, 30, 145, 402, 11, 4131, 51, 575, 32, 61, 369, 71, 66, 770, 12, 1054, 75, 100, 2198, 8, 4, 105, 37, 69, 147, 712, 75, 3543, 44, 257, 390, 5, 69, 263, 514, 105, 50, 286, 1814, 23, 4, 123, 13, 161, 40, 5, 421, 4, 116, 16, 897, 13, 2, 40, 319, 5872, 112, 6700, 11, 4803, 121, 25, 70, 3468, 4, 719, 3798, 13, 18, 31, 62, 40, 8, 7200, 4, 2, 7, 14, 123, 5, 942, 25, 8, 721, 12, 145, 5, 202, 12, 160, 580, 202, 12, 6, 52, 58, 2, 92, 401, 728, 12, 39, 14, 251, 8, 15, 251, 5, 2, 12, 38, 84, 80, 124, 12, 9, 23]),\n","       list([1, 17, 6, 194, 337, 7, 4, 204, 22, 45, 254, 8, 106, 14, 123, 4, 2, 270, 2, 5, 2, 2, 732, 2098, 101, 405, 39, 14, 1034, 4, 1310, 9, 115, 50, 305, 12, 47, 4, 168, 5, 235, 7, 38, 111, 699, 102, 7, 4, 4039, 9245, 9, 24, 6, 78, 1099, 17, 2345, 2, 21, 27, 9685, 6139, 5, 2, 1603, 92, 1183, 4, 1310, 7, 4, 204, 42, 97, 90, 35, 221, 109, 29, 127, 27, 118, 8, 97, 12, 157, 21, 6789, 2, 9, 6, 66, 78, 1099, 4, 631, 1191, 5, 2642, 272, 191, 1070, 6, 7585, 8, 2197, 2, 2, 544, 5, 383, 1271, 848, 1468, 2, 497, 2, 8, 1597, 8778, 2, 21, 60, 27, 239, 9, 43, 8368, 209, 405, 10, 10, 12, 764, 40, 4, 248, 20, 12, 16, 5, 174, 1791, 72, 7, 51, 6, 1739, 22, 4, 204, 131, 9])],\n","      dtype=object)"]},"metadata":{},"execution_count":5}]},{"cell_type":"markdown","source":["Non possiamo fornire alla rete neurale come input un elenco di numeri di lunghzza variabile per ogni recensione. La dimensione dell'input dovrebbe essere sempre la stessa. Allora trasformiamo l'input da una sequenza di numeri ad un vettore di 10.000 valori binari. La poizione i-esima del vettore sarà 1 se la parola i-esima fa parte della recensione, 0 altrimenti."],"metadata":{"id":"wK8_CwED-eyn"}},{"cell_type":"code","source":["# Definiamo una funzione che esegue la trasformazione vista sopra. Chiamiamo l'insieme di\n","# dati modificato x_train (ed x_test). L'uso di questi nomi per indicare i dati di addestramento\n","# e di test è abbastanza comune.\n","\n","def vectorize_sequences(sequences, dimension=10000):\n","  results = np.zeros((len(sequences), dimension))\n","  for i, sequence in enumerate(sequences):\n","    results[i, sequence] = 1.\n","  return results\n","\n","x_train = vectorize_sequences(imdb_train_data)\n","x_test = vectorize_sequences(imdb_test_data)"],"metadata":{"id":"U2GsOxVp8axl","executionInfo":{"status":"ok","timestamp":1647962799594,"user_tz":-60,"elapsed":3936,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}}},"execution_count":6,"outputs":[]},{"cell_type":"code","source":["x_train[0]"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"K1DTD4t-_aNS","executionInfo":{"status":"ok","timestamp":1647962801777,"user_tz":-60,"elapsed":221,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"7354e1e6-98f1-45f4-f17c-30f20c02960d"},"execution_count":7,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([0., 1., 1., ..., 0., 0., 0.])"]},"metadata":{},"execution_count":7}]},{"cell_type":"code","source":["# Per omogenità, chiamiamo le etichette di addestramento e di test con i nomi y_train e y_test\n","\n","y_train = imdb_train_labels\n","y_test = imdb_test_labels"],"metadata":{"id":"HrMKA3z9Bi_k","executionInfo":{"status":"ok","timestamp":1647962803247,"user_tz":-60,"elapsed":5,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}}},"execution_count":8,"outputs":[]},{"cell_type":"code","source":["# Dividiamo l'insieme di training in un insieme di training parziale (x_partial_train) formato dalle recensioni\n","# dalla 10.000 in poi, e in un insieme di validazione formato dalle prime 10.000 recensioni. Stessa cosa per le\n","# etichette.\n","\n","x_val = x_train[:10000]\n","x_partial_train = x_train[10000:]\n","y_val = y_train[:10000]\n","y_partial_train = y_train[10000:]"],"metadata":{"id":"wKV08auVEh0l","executionInfo":{"status":"ok","timestamp":1647962804099,"user_tz":-60,"elapsed":4,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}}},"execution_count":9,"outputs":[]},{"cell_type":"markdown","source":["# Addestramento"],"metadata":{"id":"vapD01ibS8wA"}},{"cell_type":"code","source":["# Addestriamo sull'insieme di addestramento parziale e validiamo i risultati con l'insieme di validazione\n","\n","imdb_network = tf.keras.models.Sequential([\n","   tf.keras.layers.Dense(16, activation='relu', input_shape=(10000,)),\n","   tf.keras.layers.Dense(16, activation='relu'),\n","   tf.keras.layers.Dense(1, activation='sigmoid')\n","])\n","imdb_network.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])\n","imdb_history = imdb_network.fit(x_partial_train, y_partial_train, epochs=20, batch_size=512, validation_data=(x_val, y_val))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"kMeXruIUI9WA","executionInfo":{"status":"ok","timestamp":1647962849058,"user_tz":-60,"elapsed":43470,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"33fc7413-127c-40eb-b069-8544623b5cca"},"execution_count":10,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/20\n","30/30 [==============================] - 2s 50ms/step - loss: 0.5024 - accuracy: 0.7783 - val_loss: 0.3592 - val_accuracy: 0.8746\n","Epoch 2/20\n","30/30 [==============================] - 1s 36ms/step - loss: 0.2888 - accuracy: 0.9030 - val_loss: 0.2933 - val_accuracy: 0.8890\n","Epoch 3/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.2138 - accuracy: 0.9285 - val_loss: 0.2747 - val_accuracy: 0.8912\n","Epoch 4/20\n","30/30 [==============================] - 1s 36ms/step - loss: 0.1683 - accuracy: 0.9462 - val_loss: 0.2867 - val_accuracy: 0.8862\n","Epoch 5/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.1329 - accuracy: 0.9581 - val_loss: 0.2944 - val_accuracy: 0.8847\n","Epoch 6/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.1127 - accuracy: 0.9641 - val_loss: 0.2977 - val_accuracy: 0.8851\n","Epoch 7/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.0884 - accuracy: 0.9737 - val_loss: 0.3278 - val_accuracy: 0.8779\n","Epoch 8/20\n","30/30 [==============================] - 1s 36ms/step - loss: 0.0718 - accuracy: 0.9816 - val_loss: 0.3527 - val_accuracy: 0.8799\n","Epoch 9/20\n","30/30 [==============================] - 1s 36ms/step - loss: 0.0567 - accuracy: 0.9863 - val_loss: 0.4396 - val_accuracy: 0.8608\n","Epoch 10/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.0505 - accuracy: 0.9859 - val_loss: 0.3935 - val_accuracy: 0.8790\n","Epoch 11/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.0321 - accuracy: 0.9941 - val_loss: 0.4551 - val_accuracy: 0.8669\n","Epoch 12/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.0306 - accuracy: 0.9941 - val_loss: 0.4510 - val_accuracy: 0.8750\n","Epoch 13/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.0236 - accuracy: 0.9955 - val_loss: 0.4867 - val_accuracy: 0.8706\n","Epoch 14/20\n","30/30 [==============================] - 2s 61ms/step - loss: 0.0204 - accuracy: 0.9961 - val_loss: 0.5109 - val_accuracy: 0.8727\n","Epoch 15/20\n","30/30 [==============================] - 1s 49ms/step - loss: 0.0130 - accuracy: 0.9986 - val_loss: 0.5369 - val_accuracy: 0.8704\n","Epoch 16/20\n","30/30 [==============================] - 1s 36ms/step - loss: 0.0096 - accuracy: 0.9993 - val_loss: 0.5774 - val_accuracy: 0.8700\n","Epoch 17/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.0063 - accuracy: 0.9997 - val_loss: 0.6433 - val_accuracy: 0.8642\n","Epoch 18/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.0101 - accuracy: 0.9979 - val_loss: 0.6441 - val_accuracy: 0.8682\n","Epoch 19/20\n","30/30 [==============================] - 1s 36ms/step - loss: 0.0029 - accuracy: 0.9999 - val_loss: 0.6846 - val_accuracy: 0.8676\n","Epoch 20/20\n","30/30 [==============================] - 1s 37ms/step - loss: 0.0060 - accuracy: 0.9989 - val_loss: 0.6956 - val_accuracy: 0.8670\n"]}]},{"cell_type":"code","source":["imdb_network.evaluate(x_test, y_test)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Cc_UVTyGH_Yb","executionInfo":{"status":"ok","timestamp":1647962857488,"user_tz":-60,"elapsed":3001,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"ea81c635-33ff-49ee-b38f-41c3d181203d"},"execution_count":11,"outputs":[{"output_type":"stream","name":"stdout","text":["782/782 [==============================] - 2s 3ms/step - loss: 0.7639 - accuracy: 0.8524\n"]},{"output_type":"execute_result","data":{"text/plain":["[0.7639040350914001, 0.8523600101470947]"]},"metadata":{},"execution_count":11}]},{"cell_type":"code","source":["# Definiamo una funzione per visualizzare l'andamento dell'errore e dell'accuratezza sia\n","# per l'insieme di addestramento che per quello di validazione\n","\n","def display_loss_and_accuracy_withval(history):\n","  history_dict = history.history\n","\n","  loss_values = history_dict['loss']\n","  val_loss_values = history_dict['val_loss']\n","  acc_values = history_dict['accuracy']\n","  val_acc_values = history_dict['val_accuracy']\n","  epochs = range(1, len(loss_values) + 1)\n","\n","  plt.plot(epochs, loss_values, 'bo', label='Training loss')\n","  plt.plot(epochs, val_loss_values, 'b', label='Validation loss')\n","  plt.title('Training and validation loss')\n","  plt.xlabel('Epochs')\n","  plt.ylabel('Loss')\n","  plt.legend()\n","  plt.show()\n","\n","  plt.plot(epochs, acc_values, 'bo', label='Training accuracy')\n","  plt.plot(epochs, val_acc_values, 'b', label='Validation accuracy')\n","  plt.title('Training and validation accuracy')\n","  plt.xlabel('Epochs')\n","  plt.ylabel('Loss')\n","  plt.legend()\n","  plt.show()"],"metadata":{"id":"T8kKUff5EQnV","executionInfo":{"status":"ok","timestamp":1647962860357,"user_tz":-60,"elapsed":218,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}}},"execution_count":12,"outputs":[]},{"cell_type":"code","source":["display_loss_and_accuracy_withval(imdb_history)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":573},"id":"cC6fGtfcF8gb","executionInfo":{"status":"ok","timestamp":1647962863447,"user_tz":-60,"elapsed":663,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"e6e810a0-39ac-4f4c-ef6b-2bb630d840b8"},"execution_count":13,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["Una volta individuata l'epoca migliore (la numero 3), riaddestriamo la rete usando l'insieme di training completo, e valutiamo l'accuratezza sull'insieme di test."],"metadata":{"id":"rx1Cu88oPi7F"}},{"cell_type":"code","source":["imdb_network_full = tf.keras.models.Sequential([\n","   tf.keras.layers.Dense(16, activation='relu', input_shape=(10000,)),\n","   tf.keras.layers.Dense(16, activation='relu'),\n","   tf.keras.layers.Dense(1, activation='sigmoid')\n","])\n","imdb_network_full.compile (optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])\n","imdb_history_full = imdb_network_full.fit(x_train, y_train, epochs=3, batch_size=512)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"M0jKzyoGKMIJ","executionInfo":{"status":"ok","timestamp":1647962911872,"user_tz":-60,"elapsed":5593,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"1af40c0a-261a-45c2-80a1-2e93467b880f"},"execution_count":15,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/3\n","49/49 [==============================] - 2s 30ms/step - loss: 0.4589 - accuracy: 0.8103\n","Epoch 2/3\n","49/49 [==============================] - 1s 29ms/step - loss: 0.2581 - accuracy: 0.9076\n","Epoch 3/3\n","49/49 [==============================] - 1s 28ms/step - loss: 0.2008 - accuracy: 0.9270\n"]}]},{"cell_type":"code","source":["imdb_network_full.evaluate(x_test, y_test)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"PhaTtthpKYFL","executionInfo":{"status":"ok","timestamp":1647962920202,"user_tz":-60,"elapsed":2887,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"1c7daa86-ed62-416e-8765-a9f41055e84a"},"execution_count":16,"outputs":[{"output_type":"stream","name":"stdout","text":["782/782 [==============================] - 2s 3ms/step - loss: 0.2936 - accuracy: 0.8830\n"]},{"output_type":"execute_result","data":{"text/plain":["[0.29359811544418335, 0.8830000162124634]"]},"metadata":{},"execution_count":16}]},{"cell_type":"markdown","source":["# Ridurre l'overfitting"],"metadata":{"id":"-SFWcms6SAmG"}},{"cell_type":"markdown","source":["Vediamo alcune tecniche standard per cercare di ridurre l'overfitting."],"metadata":{"id":"BPt48RwUFkDQ"}},{"cell_type":"markdown","source":["## Semplificare la rete"],"metadata":{"id":"6LfUso6KW_zE"}},{"cell_type":"markdown","source":["Il modo più immediato per ridurre l'overfitting è semplificare la rete neurale. L'overfitting è causato dal fatto che la rete è così complessa che riesce a \"imparare a memoria\" le caratteristiche peculiari dell'insieme di dati di addestramento, e non ha bisogno di \"generalizzare\". Proviamo con una nuova rete con strati interni di 4 neuroni."],"metadata":{"id":"kmW3h3nlTe2e"}},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"C63taSScLBjV","executionInfo":{"status":"ok","timestamp":1647962955471,"user_tz":-60,"elapsed":21680,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"73a70a15-52e5-4f99-ec4d-5486dfb1439e"},"source":["imdb_network2 = tf.keras.models.Sequential([\n","   tf.keras.layers.Dense(4, activation='relu', input_shape=(10000,)),\n","   tf.keras.layers.Dense(4, activation='relu'),\n","   tf.keras.layers.Dense(1, activation='sigmoid')\n","])\n","imdb_network2.compile (optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])\n","imdb_history2 = imdb_network2.fit(\n","  x_partial_train,\n","  y_partial_train,\n","  epochs=20,\n","  batch_size=512,\n","  validation_data=(x_val, y_val))"],"execution_count":17,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/20\n","30/30 [==============================] - 2s 43ms/step - loss: 0.6502 - accuracy: 0.7247 - val_loss: 0.6003 - val_accuracy: 0.7952\n","Epoch 2/20\n","30/30 [==============================] - 1s 34ms/step - loss: 0.5462 - accuracy: 0.8459 - val_loss: 0.5098 - val_accuracy: 0.8469\n","Epoch 3/20\n","30/30 [==============================] - 1s 34ms/step - loss: 0.4515 - accuracy: 0.8797 - val_loss: 0.4330 - val_accuracy: 0.8655\n","Epoch 4/20\n","30/30 [==============================] - 1s 35ms/step - loss: 0.3731 - accuracy: 0.8969 - val_loss: 0.3747 - val_accuracy: 0.8738\n","Epoch 5/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.3121 - accuracy: 0.9070 - val_loss: 0.3336 - val_accuracy: 0.8830\n","Epoch 6/20\n","30/30 [==============================] - 1s 34ms/step - loss: 0.2669 - accuracy: 0.9175 - val_loss: 0.3069 - val_accuracy: 0.8875\n","Epoch 7/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.2323 - accuracy: 0.9275 - val_loss: 0.2912 - val_accuracy: 0.8881\n","Epoch 8/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.2043 - accuracy: 0.9355 - val_loss: 0.2802 - val_accuracy: 0.8921\n","Epoch 9/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.1828 - accuracy: 0.9415 - val_loss: 0.2754 - val_accuracy: 0.8927\n","Epoch 10/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.1641 - accuracy: 0.9481 - val_loss: 0.2716 - val_accuracy: 0.8923\n","Epoch 11/20\n","30/30 [==============================] - 1s 32ms/step - loss: 0.1474 - accuracy: 0.9536 - val_loss: 0.2745 - val_accuracy: 0.8898\n","Epoch 12/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.1328 - accuracy: 0.9591 - val_loss: 0.2767 - val_accuracy: 0.8906\n","Epoch 13/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.1207 - accuracy: 0.9628 - val_loss: 0.2825 - val_accuracy: 0.8899\n","Epoch 14/20\n","30/30 [==============================] - 1s 34ms/step - loss: 0.1089 - accuracy: 0.9678 - val_loss: 0.2938 - val_accuracy: 0.8876\n","Epoch 15/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.0990 - accuracy: 0.9705 - val_loss: 0.2986 - val_accuracy: 0.8878\n","Epoch 16/20\n","30/30 [==============================] - 1s 34ms/step - loss: 0.0896 - accuracy: 0.9737 - val_loss: 0.3085 - val_accuracy: 0.8866\n","Epoch 17/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.0805 - accuracy: 0.9779 - val_loss: 0.3207 - val_accuracy: 0.8842\n","Epoch 18/20\n","30/30 [==============================] - 1s 33ms/step - loss: 0.0730 - accuracy: 0.9803 - val_loss: 0.3367 - val_accuracy: 0.8820\n","Epoch 19/20\n","30/30 [==============================] - 1s 34ms/step - loss: 0.0658 - accuracy: 0.9829 - val_loss: 0.3454 - val_accuracy: 0.8834\n","Epoch 20/20\n","30/30 [==============================] - 1s 34ms/step - loss: 0.0591 - accuracy: 0.9855 - val_loss: 0.3618 - val_accuracy: 0.8804\n"]}]},{"cell_type":"code","metadata":{"id":"jTa1EKqsPssQ","executionInfo":{"status":"ok","timestamp":1647962993729,"user_tz":-60,"elapsed":325,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}}},"source":["# Definiziamo una funzione che confronta l'andamento della funzione di loss\n","# e dell'accuratezza al variare delle epoche tra il modello originale (rosso)\n","# e quello nuovo (blu).\n","\n","def display_network_comparison(original, new):\n","  epochs = range(1, len(original.history['val_loss']) + 1)\n","\n","  plt.plot(epochs, original.history['val_loss'], 'r', label='Original model')\n","  plt.plot(epochs, new.history['val_loss'], 'b', label='New model')\n","\n","  plt.title('Validation loss')\n","  plt.xlabel('Epochs')\n","  plt.ylabel('Loss')\n","  plt.legend()\n","  plt.show()\n","\n","  plt.plot(epochs, original.history['val_accuracy'], 'r', label='Original model')\n","  plt.plot(epochs, new.history['val_accuracy'], 'b', label='New model')\n","\n","  plt.title('Validation accuracy')\n","  plt.xlabel('Epochs')\n","  plt.ylabel('accuracy')\n","  plt.legend()\n","  plt.show()"],"execution_count":18,"outputs":[]},{"cell_type":"code","source":["display_network_comparison(imdb_history, imdb_history2)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":573},"id":"t4ffbojAWEG2","executionInfo":{"status":"ok","timestamp":1647962996625,"user_tz":-60,"elapsed":597,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"2c4cf644-7a66-4dc1-d222-544021f4f44d"},"execution_count":19,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["Vediamo che la rete più semplice va in overfitting dopo e degrada più lentamente. Se, di contro, rendiamo la rete più complessa, il comportamento è piuttosto strano."],"metadata":{"id":"mypVjeSqUr5s"}},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"nMLR0IZ7QUOn","executionInfo":{"status":"ok","timestamp":1647963431349,"user_tz":-60,"elapsed":179514,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"980ff1a5-845e-409b-d8f5-73422bdb7b7a"},"source":["imdb_network3 = tf.keras.models.Sequential([\n","   tf.keras.layers.Dense(512, activation='relu', input_shape=(10000,)),\n","   tf.keras.layers.Dense(512, activation='relu'),\n","   tf.keras.layers.Dense(1, activation='sigmoid')\n","])\n","imdb_network3.compile (optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])\n","imdb_history3 = imdb_network3.fit(\n","  x_partial_train,\n","  y_partial_train,\n","  epochs=20,\n","  batch_size=512,\n","  validation_data=(x_val, y_val))"],"execution_count":22,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/20\n","30/30 [==============================] - 10s 322ms/step - loss: 0.5479 - accuracy: 0.7709 - val_loss: 0.3609 - val_accuracy: 0.8451\n","Epoch 2/20\n","30/30 [==============================] - 9s 297ms/step - loss: 0.2581 - accuracy: 0.8936 - val_loss: 0.3262 - val_accuracy: 0.8623\n","Epoch 3/20\n","30/30 [==============================] - 9s 296ms/step - loss: 0.1444 - accuracy: 0.9472 - val_loss: 0.3369 - val_accuracy: 0.8681\n","Epoch 4/20\n","30/30 [==============================] - 9s 298ms/step - loss: 0.0606 - accuracy: 0.9810 - val_loss: 0.3834 - val_accuracy: 0.8899\n","Epoch 5/20\n","30/30 [==============================] - 9s 297ms/step - loss: 0.1043 - accuracy: 0.9802 - val_loss: 0.3582 - val_accuracy: 0.8878\n","Epoch 6/20\n","30/30 [==============================] - 9s 294ms/step - loss: 0.0028 - accuracy: 0.9999 - val_loss: 0.5215 - val_accuracy: 0.8832\n","Epoch 7/20\n","30/30 [==============================] - 9s 298ms/step - loss: 3.2488e-04 - accuracy: 1.0000 - val_loss: 0.6133 - val_accuracy: 0.8858\n","Epoch 8/20\n","30/30 [==============================] - 9s 300ms/step - loss: 5.2693e-05 - accuracy: 1.0000 - val_loss: 0.7133 - val_accuracy: 0.8852\n","Epoch 9/20\n","30/30 [==============================] - 9s 297ms/step - loss: 9.8819e-06 - accuracy: 1.0000 - val_loss: 0.8129 - val_accuracy: 0.8849\n","Epoch 10/20\n","30/30 [==============================] - 9s 297ms/step - loss: 2.3700e-06 - accuracy: 1.0000 - val_loss: 1.0992 - val_accuracy: 0.8681\n","Epoch 11/20\n","30/30 [==============================] - 9s 294ms/step - loss: 0.3114 - accuracy: 0.9755 - val_loss: 0.6001 - val_accuracy: 0.8871\n","Epoch 12/20\n","30/30 [==============================] - 9s 297ms/step - loss: 3.2205e-05 - accuracy: 1.0000 - val_loss: 0.6318 - val_accuracy: 0.8868\n","Epoch 13/20\n","30/30 [==============================] - 9s 295ms/step - loss: 1.6237e-05 - accuracy: 1.0000 - val_loss: 0.6666 - val_accuracy: 0.8869\n","Epoch 14/20\n","30/30 [==============================] - 9s 298ms/step - loss: 8.2815e-06 - accuracy: 1.0000 - val_loss: 0.7125 - val_accuracy: 0.8869\n","Epoch 15/20\n","30/30 [==============================] - 9s 296ms/step - loss: 3.7071e-06 - accuracy: 1.0000 - val_loss: 0.7787 - val_accuracy: 0.8870\n","Epoch 16/20\n","30/30 [==============================] - 9s 300ms/step - loss: 1.4230e-06 - accuracy: 1.0000 - val_loss: 0.8574 - val_accuracy: 0.8872\n","Epoch 17/20\n","30/30 [==============================] - 9s 302ms/step - loss: 5.1974e-07 - accuracy: 1.0000 - val_loss: 0.9408 - val_accuracy: 0.8872\n","Epoch 18/20\n","30/30 [==============================] - 9s 299ms/step - loss: 1.9272e-07 - accuracy: 1.0000 - val_loss: 1.0142 - val_accuracy: 0.8873\n","Epoch 19/20\n","30/30 [==============================] - 9s 298ms/step - loss: 7.9524e-08 - accuracy: 1.0000 - val_loss: 1.0715 - val_accuracy: 0.8871\n","Epoch 20/20\n","30/30 [==============================] - 9s 297ms/step - loss: 3.8117e-08 - accuracy: 1.0000 - val_loss: 1.1134 - val_accuracy: 0.8875\n"]}]},{"cell_type":"code","source":["display_network_comparison(imdb_history, imdb_history3)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":573},"id":"0trs2kTxJx0N","executionInfo":{"status":"ok","timestamp":1647963717290,"user_tz":-60,"elapsed":834,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"26eb8a69-b59f-4d98-80db-f955a20c3fd7"},"execution_count":23,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["## Regolarizzare i pesi"],"metadata":{"id":"NcKt38dmXDH-"}},{"cell_type":"markdown","source":["Un altro modo per ridurre l'overfitting è quello di cercare di ottenere reti con pesi piccoli (più vicini a 0). Pesi molto grandi tendono infatti a rendere la rete altamente sensibile a piccole variazioni degli input.\n","\n","Per realizzare ciò, si può modificare la funzione di loss in modo da aggiungere un termine che dipenda dal valore dei pesi. Tipicamente si parla di:\n","\n","*   Regolarizzazione L1: il costo aggiunto è proporzionale alla somma dei valori assoluti dei pesi\n","*   Regolarizzazione L2: il costo aggiunto è proporzionale alla somma dei quadrati dei pesi\n","\n","In Keras, la regolarizzazione si ottiene aggiungendo il parametro `kernel_regularizer` agli strati della rete neurale.\n","\n","\n","\n"],"metadata":{"id":"bUUx6obaXRIM"}},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"id":"SgkcU0mpWL3z","executionInfo":{"status":"ok","timestamp":1647963866692,"user_tz":-60,"elapsed":43241,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"f7a92574-3c34-462b-eb51-aa573f380acc"},"source":["# Ecco la stessa rete con regolarizzazione L2. Il valore 0.001 è il coefficiente per cui moltiplicare\n","# la somma dei pesi al quadrato prima di aggiungerlo alla funzione di loss. Valori elevati danno più\n","# importanza ai pesi rispetto che alla funzione di loss originale, viceversa per valori piccoli.\n","\n","imdb_network4 = tf.keras.models.Sequential([\n","   tf.keras.layers.Dense(16, activation='relu', kernel_regularizer=tf.keras.regularizers.l2(0.001), input_shape=(10000,)),\n","   tf.keras.layers.Dense(16, activation='relu', kernel_regularizer=tf.keras.regularizers.l2(0.001)),\n","   tf.keras.layers.Dense(1, activation='sigmoid')\n","])\n","imdb_network4.compile (optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])\n","imdb_history4 = imdb_network4.fit(\n","  x_partial_train,\n","  y_partial_train,\n","  epochs=20,\n","  batch_size=512,\n","  validation_data=(x_val, y_val))\n","\n","display_network_comparison(imdb_history, imdb_history4)"],"execution_count":24,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/20\n","30/30 [==============================] - 3s 59ms/step - loss: 0.5598 - accuracy: 0.7919 - val_loss: 0.4660 - val_accuracy: 0.8280\n","Epoch 2/20\n","30/30 [==============================] - 1s 41ms/step - loss: 0.3701 - accuracy: 0.8991 - val_loss: 0.3624 - val_accuracy: 0.8870\n","Epoch 3/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.3008 - accuracy: 0.9181 - val_loss: 0.3430 - val_accuracy: 0.8850\n","Epoch 4/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.2627 - accuracy: 0.9306 - val_loss: 0.3294 - val_accuracy: 0.8900\n","Epoch 5/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.2426 - accuracy: 0.9367 - val_loss: 0.3413 - val_accuracy: 0.8825\n","Epoch 6/20\n","30/30 [==============================] - 1s 41ms/step - loss: 0.2288 - accuracy: 0.9419 - val_loss: 0.3787 - val_accuracy: 0.8716\n","Epoch 7/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.2135 - accuracy: 0.9510 - val_loss: 0.3464 - val_accuracy: 0.8845\n","Epoch 8/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.2081 - accuracy: 0.9506 - val_loss: 0.3463 - val_accuracy: 0.8834\n","Epoch 9/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.1958 - accuracy: 0.9563 - val_loss: 0.3965 - val_accuracy: 0.8712\n","Epoch 10/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.1888 - accuracy: 0.9599 - val_loss: 0.4128 - val_accuracy: 0.8603\n","Epoch 11/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1869 - accuracy: 0.9597 - val_loss: 0.3677 - val_accuracy: 0.8810\n","Epoch 12/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1805 - accuracy: 0.9618 - val_loss: 0.3730 - val_accuracy: 0.8804\n","Epoch 13/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.1765 - accuracy: 0.9629 - val_loss: 0.4143 - val_accuracy: 0.8638\n","Epoch 14/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1737 - accuracy: 0.9653 - val_loss: 0.3887 - val_accuracy: 0.8782\n","Epoch 15/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.1633 - accuracy: 0.9688 - val_loss: 0.4683 - val_accuracy: 0.8583\n","Epoch 16/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1657 - accuracy: 0.9671 - val_loss: 0.4121 - val_accuracy: 0.8706\n","Epoch 17/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.1651 - accuracy: 0.9677 - val_loss: 0.4063 - val_accuracy: 0.8752\n","Epoch 18/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1548 - accuracy: 0.9723 - val_loss: 0.4154 - val_accuracy: 0.8735\n","Epoch 19/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.1562 - accuracy: 0.9700 - val_loss: 0.4169 - val_accuracy: 0.8749\n","Epoch 20/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.1565 - accuracy: 0.9688 - val_loss: 0.4462 - val_accuracy: 0.8669\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["## Dropout\n","\n","Il *dropout* consiste nel mettere a zero, in maniera casuale, alcuni degli output di uno strato intermedio. Il *valore di dropout* è la percentuale di output da azzerare.\n","\n","Il dropout sembra una tecnica bizzarra... l'idea è che mettendo casualmente a zero alcuni output gli strati succesivi non possono fare troppo affidamento a nessun neurone specifico di quelli precedenti, ma devono cercare di sfruttare tutte le informazioni disponibili."],"metadata":{"id":"vS6hDchUYkwI"}},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"id":"K0KDH2EdZV8x","executionInfo":{"status":"ok","timestamp":1647964072232,"user_tz":-60,"elapsed":42425,"user":{"displayName":"Gianluca Amato","photoUrl":"https://lh3.googleusercontent.com/a-/AOh14GhEmlMukD8qDnxG1PlDxK7Dzm1rjUKZSz0BhiJc9w=s64","userId":"18269286707108730791"}},"outputId":"e5f3ed77-9d7f-4eeb-8791-5b946cfba3ce"},"source":["imdb_network5 = tf.keras.models.Sequential([\n","   tf.keras.layers.Dense(16, activation='relu', input_shape=(10000,)),\n","   tf.keras.layers.Dropout(0.5),\n","   tf.keras.layers.Dense(16, activation='relu'),\n","   tf.keras.layers.Dropout(0.5),\n","   tf.keras.layers.Dense(1, activation='sigmoid')\n","])\n","imdb_network5.compile (optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])\n","imdb_history5 = imdb_network5.fit(\n","  x_partial_train,\n","  y_partial_train,\n","  epochs=20,\n","  batch_size=512,\n","  validation_data=(x_val, y_val))\n","\n","display_network_comparison(imdb_history, imdb_history5)"],"execution_count":25,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/20\n","30/30 [==============================] - 2s 51ms/step - loss: 0.6299 - accuracy: 0.6318 - val_loss: 0.5114 - val_accuracy: 0.8486\n","Epoch 2/20\n","30/30 [==============================] - 2s 60ms/step - loss: 0.5030 - accuracy: 0.7605 - val_loss: 0.3950 - val_accuracy: 0.8745\n","Epoch 3/20\n","30/30 [==============================] - 2s 70ms/step - loss: 0.4242 - accuracy: 0.8120 - val_loss: 0.3512 - val_accuracy: 0.8722\n","Epoch 4/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.3625 - accuracy: 0.8589 - val_loss: 0.3008 - val_accuracy: 0.8885\n","Epoch 5/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.3123 - accuracy: 0.8832 - val_loss: 0.2863 - val_accuracy: 0.8867\n","Epoch 6/20\n","30/30 [==============================] - 1s 42ms/step - loss: 0.2793 - accuracy: 0.9005 - val_loss: 0.2732 - val_accuracy: 0.8908\n","Epoch 7/20\n","30/30 [==============================] - 1s 42ms/step - loss: 0.2451 - accuracy: 0.9195 - val_loss: 0.2757 - val_accuracy: 0.8905\n","Epoch 8/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.2147 - accuracy: 0.9296 - val_loss: 0.2899 - val_accuracy: 0.8887\n","Epoch 9/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1896 - accuracy: 0.9393 - val_loss: 0.2941 - val_accuracy: 0.8864\n","Epoch 10/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1708 - accuracy: 0.9464 - val_loss: 0.3088 - val_accuracy: 0.8850\n","Epoch 11/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1587 - accuracy: 0.9488 - val_loss: 0.3301 - val_accuracy: 0.8868\n","Epoch 12/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1353 - accuracy: 0.9567 - val_loss: 0.3448 - val_accuracy: 0.8880\n","Epoch 13/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1211 - accuracy: 0.9607 - val_loss: 0.3748 - val_accuracy: 0.8861\n","Epoch 14/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.1151 - accuracy: 0.9625 - val_loss: 0.3970 - val_accuracy: 0.8847\n","Epoch 15/20\n","30/30 [==============================] - 1s 39ms/step - loss: 0.1094 - accuracy: 0.9657 - val_loss: 0.4178 - val_accuracy: 0.8849\n","Epoch 16/20\n","30/30 [==============================] - 1s 42ms/step - loss: 0.1005 - accuracy: 0.9679 - val_loss: 0.4479 - val_accuracy: 0.8848\n","Epoch 17/20\n","30/30 [==============================] - 1s 41ms/step - loss: 0.0901 - accuracy: 0.9707 - val_loss: 0.4932 - val_accuracy: 0.8851\n","Epoch 18/20\n","30/30 [==============================] - 1s 42ms/step - loss: 0.0829 - accuracy: 0.9743 - val_loss: 0.4953 - val_accuracy: 0.8827\n","Epoch 19/20\n","30/30 [==============================] - 1s 40ms/step - loss: 0.0820 - accuracy: 0.9739 - val_loss: 0.5254 - val_accuracy: 0.8859\n","Epoch 20/20\n","30/30 [==============================] - 1s 42ms/step - loss: 0.0797 - accuracy: 0.9747 - val_loss: 0.5694 - val_accuracy: 0.8851\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]}]}