 Ver em TensorFlow.org
|
 Executar em Google Colab
|
 Ver código fonte no GitHub
|
 Baixar notebook
|
Este tutorial fornece um exemplo de carregamento de dados de matrizes NumPy para um tf.data.Dataset.
Este exemplo carrega o conjunto de dados MNIST de um arquivo .npz. No entanto, a fonte das matrizes NumPy não é importante.
Configuração
try:
# %tensorflow_version only exists in Colab.
%tensorflow_version 2.x
except Exception:
pass
from __future__ import absolute_import, division, print_function, unicode_literals
import numpy as np
import tensorflow as tf
Carregar um arquivo .npz
DATA_URL = 'https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz'
path = tf.keras.utils.get_file('mnist.npz', DATA_URL)
with np.load(path) as data:
train_examples = data['x_train']
train_labels = data['y_train']
test_examples = data['x_test']
test_labels = data['y_test']
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz 8192/11490434 [..............................] - ETA: 0s 4202496/11490434 [=========>....................] - ETA: 0s 11493376/11490434 [==============================] - 0s 0us/step
Carregar matrizes NumPy com tf.data.Dataset
Supondo que você tenha uma matriz de exemplos e uma matriz correspondente de rótulos, passe as duas matrizes como uma tupla para tf.data.Dataset.from_tensor_slices para criar um tf.data.Dataset.
train_dataset = tf.data.Dataset.from_tensor_slices((train_examples, train_labels))
test_dataset = tf.data.Dataset.from_tensor_slices((test_examples, test_labels))
Usar o conjunto de dados
Aleatório e lote dos conjuntos de dados
BATCH_SIZE = 64
SHUFFLE_BUFFER_SIZE = 100
train_dataset = train_dataset.shuffle(SHUFFLE_BUFFER_SIZE).batch(BATCH_SIZE)
test_dataset = test_dataset.batch(BATCH_SIZE)
Construir e treinar um modelo
model = tf.keras.Sequential([
tf.keras.layers.Flatten(input_shape=(28, 28)),
tf.keras.layers.Dense(128, activation='relu'),
tf.keras.layers.Dense(10)
])
model.compile(optimizer=tf.keras.optimizers.RMSprop(),
loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
metrics=['sparse_categorical_accuracy'])
model.fit(train_dataset, epochs=10)
Epoch 1/10 1/938 [..............................] - ETA: 0s - loss: 172.8731 - sparse_categorical_accuracy: 0.0312 30/938 [..............................] - ETA: 1s - loss: 37.2118 - sparse_categorical_accuracy: 0.5885 60/938 [>.............................] - ETA: 1s - loss: 23.3260 - sparse_categorical_accuracy: 0.7026 91/938 [=>............................] - ETA: 1s - loss: 18.1632 - sparse_categorical_accuracy: 0.7455 121/938 [==>...........................] - ETA: 1s - loss: 15.1468 - sparse_categorical_accuracy: 0.7727 152/938 [===>..........................] - ETA: 1s - loss: 13.4535 - sparse_categorical_accuracy: 0.7880 182/938 [====>.........................] - ETA: 1s - loss: 11.8808 - sparse_categorical_accuracy: 0.8036 212/938 [=====>........................] - ETA: 1s - loss: 10.8656 - sparse_categorical_accuracy: 0.8135 242/938 [======>.......................] - ETA: 1s - loss: 10.0613 - sparse_categorical_accuracy: 0.8198 273/938 [=======>......................] - ETA: 1s - loss: 9.3106 - sparse_categorical_accuracy: 0.8266 303/938 [========>.....................] - ETA: 1s - loss: 8.6906 - sparse_categorical_accuracy: 0.8332 334/938 [=========>....................] - ETA: 1s - loss: 8.1657 - sparse_categorical_accuracy: 0.8388 363/938 [==========>...................] - ETA: 0s - loss: 7.7063 - sparse_categorical_accuracy: 0.8437 394/938 [===========>..................] - ETA: 0s - loss: 7.3061 - sparse_categorical_accuracy: 0.8482 425/938 [============>.................] - ETA: 0s - loss: 6.9671 - sparse_categorical_accuracy: 0.8514 455/938 [=============>................] - ETA: 0s - loss: 6.6436 - sparse_categorical_accuracy: 0.8552 484/938 [==============>...............] - ETA: 0s - loss: 6.3633 - sparse_categorical_accuracy: 0.8577 514/938 [===============>..............] - ETA: 0s - loss: 6.1072 - sparse_categorical_accuracy: 0.8596 543/938 [================>.............] - ETA: 0s - loss: 5.8573 - sparse_categorical_accuracy: 0.8629 573/938 [=================>............] - ETA: 0s - loss: 5.6450 - sparse_categorical_accuracy: 0.8653 603/938 [==================>...........] - ETA: 0s - loss: 5.4387 - sparse_categorical_accuracy: 0.8672 633/938 [===================>..........] - ETA: 0s - loss: 5.2480 - sparse_categorical_accuracy: 0.8694 663/938 [====================>.........] - ETA: 0s - loss: 5.0754 - sparse_categorical_accuracy: 0.8709 693/938 [=====================>........] - ETA: 0s - loss: 4.9066 - sparse_categorical_accuracy: 0.8726 722/938 [======================>.......] - ETA: 0s - loss: 4.7540 - sparse_categorical_accuracy: 0.8733 752/938 [=======================>......] - ETA: 0s - loss: 4.6059 - sparse_categorical_accuracy: 0.8740 782/938 [========================>.....] - ETA: 0s - loss: 4.4732 - sparse_categorical_accuracy: 0.8745 812/938 [========================>.....] - ETA: 0s - loss: 4.3439 - sparse_categorical_accuracy: 0.8755 842/938 [=========================>....] - ETA: 0s - loss: 4.2239 - sparse_categorical_accuracy: 0.8764 873/938 [==========================>...] - ETA: 0s - loss: 4.0983 - sparse_categorical_accuracy: 0.8776 903/938 [===========================>..] - ETA: 0s - loss: 3.9835 - sparse_categorical_accuracy: 0.8790 934/938 [============================>.] - ETA: 0s - loss: 3.8679 - sparse_categorical_accuracy: 0.8810 938/938 [==============================] - 2s 2ms/step - loss: 3.8568 - sparse_categorical_accuracy: 0.8812 Epoch 2/10 1/938 [..............................] - ETA: 0s - loss: 0.6360 - sparse_categorical_accuracy: 0.9219 31/938 [..............................] - ETA: 1s - loss: 0.9947 - sparse_categorical_accuracy: 0.9002 61/938 [>.............................] - ETA: 1s - loss: 0.8230 - sparse_categorical_accuracy: 0.9106 90/938 [=>............................] - ETA: 1s - loss: 0.7770 - sparse_categorical_accuracy: 0.9106 120/938 [==>...........................] - ETA: 1s - loss: 0.7505 - sparse_categorical_accuracy: 0.9107 151/938 [===>..........................] - ETA: 1s - loss: 0.7698 - sparse_categorical_accuracy: 0.9088 180/938 [====>.........................] - ETA: 1s - loss: 0.7313 - sparse_categorical_accuracy: 0.9116 209/938 [=====>........................] - ETA: 1s - loss: 0.7289 - sparse_categorical_accuracy: 0.9108 238/938 [======>.......................] - ETA: 1s - loss: 0.7260 - sparse_categorical_accuracy: 0.9117 267/938 [=======>......................] - ETA: 1s - loss: 0.7111 - sparse_categorical_accuracy: 0.9122 296/938 [========>.....................] - ETA: 1s - loss: 0.6884 - sparse_categorical_accuracy: 0.9135 326/938 [=========>....................] - ETA: 1s - loss: 0.6828 - sparse_categorical_accuracy: 0.9148 356/938 [==========>...................] - ETA: 0s - loss: 0.6689 - sparse_categorical_accuracy: 0.9148 386/938 [===========>..................] - ETA: 0s - loss: 0.6592 - sparse_categorical_accuracy: 0.9153 415/938 [============>.................] - ETA: 0s - loss: 0.6511 - sparse_categorical_accuracy: 0.9160 445/938 [=============>................] - ETA: 0s - loss: 0.6470 - sparse_categorical_accuracy: 0.9163 474/938 [==============>...............] - ETA: 0s - loss: 0.6420 - sparse_categorical_accuracy: 0.9165 504/938 [===============>..............] - ETA: 0s - loss: 0.6399 - sparse_categorical_accuracy: 0.9166 534/938 [================>.............] - ETA: 0s - loss: 0.6296 - sparse_categorical_accuracy: 0.9180 563/938 [=================>............] - ETA: 0s - loss: 0.6224 - sparse_categorical_accuracy: 0.9184 593/938 [=================>............] - ETA: 0s - loss: 0.6200 - sparse_categorical_accuracy: 0.9187 622/938 [==================>...........] - ETA: 0s - loss: 0.6197 - sparse_categorical_accuracy: 0.9191 651/938 [===================>..........] - ETA: 0s - loss: 0.6119 - sparse_categorical_accuracy: 0.9195 680/938 [====================>.........] - ETA: 0s - loss: 0.6098 - sparse_categorical_accuracy: 0.9199 710/938 [=====================>........] - ETA: 0s - loss: 0.6063 - sparse_categorical_accuracy: 0.9204 739/938 [======================>.......] - ETA: 0s - loss: 0.6003 - sparse_categorical_accuracy: 0.9205 768/938 [=======================>......] - ETA: 0s - loss: 0.5979 - sparse_categorical_accuracy: 0.9209 797/938 [========================>.....] - ETA: 0s - loss: 0.6005 - sparse_categorical_accuracy: 0.9209 827/938 [=========================>....] - ETA: 0s - loss: 0.5947 - sparse_categorical_accuracy: 0.9217 857/938 [==========================>...] - ETA: 0s - loss: 0.5905 - sparse_categorical_accuracy: 0.9222 887/938 [===========================>..] - ETA: 0s - loss: 0.5832 - sparse_categorical_accuracy: 0.9230 916/938 [============================>.] - ETA: 0s - loss: 0.5734 - sparse_categorical_accuracy: 0.9239 938/938 [==============================] - 2s 2ms/step - loss: 0.5658 - sparse_categorical_accuracy: 0.9249 Epoch 3/10 1/938 [..............................] - ETA: 0s - loss: 0.1601 - sparse_categorical_accuracy: 0.9531 30/938 [..............................] - ETA: 1s - loss: 0.4933 - sparse_categorical_accuracy: 0.9359 60/938 [>.............................] - ETA: 1s - loss: 0.4173 - sparse_categorical_accuracy: 0.9453 89/938 [=>............................] - ETA: 1s - loss: 0.4013 - sparse_categorical_accuracy: 0.9450 119/938 [==>...........................] - ETA: 1s - loss: 0.4006 - sparse_categorical_accuracy: 0.9433 148/938 [===>..........................] - ETA: 1s - loss: 0.4278 - sparse_categorical_accuracy: 0.9399 178/938 [====>.........................] - ETA: 1s - loss: 0.4079 - sparse_categorical_accuracy: 0.9410 208/938 [=====>........................] - ETA: 1s - loss: 0.4235 - sparse_categorical_accuracy: 0.9396 238/938 [======>.......................] - ETA: 1s - loss: 0.4234 - sparse_categorical_accuracy: 0.9405 268/938 [=======>......................] - ETA: 1s - loss: 0.4201 - sparse_categorical_accuracy: 0.9401 298/938 [========>.....................] - ETA: 1s - loss: 0.4173 - sparse_categorical_accuracy: 0.9405 327/938 [=========>....................] - ETA: 1s - loss: 0.4183 - sparse_categorical_accuracy: 0.9412 357/938 [==========>...................] - ETA: 0s - loss: 0.4139 - sparse_categorical_accuracy: 0.9413 387/938 [===========>..................] - ETA: 0s - loss: 0.4160 - sparse_categorical_accuracy: 0.9417 418/938 [============>.................] - ETA: 0s - loss: 0.4131 - sparse_categorical_accuracy: 0.9421 449/938 [=============>................] - ETA: 0s - loss: 0.4133 - sparse_categorical_accuracy: 0.9421 479/938 [==============>...............] - ETA: 0s - loss: 0.4094 - sparse_categorical_accuracy: 0.9423 509/938 [===============>..............] - ETA: 0s - loss: 0.4148 - sparse_categorical_accuracy: 0.9419 538/938 [================>.............] - ETA: 0s - loss: 0.4124 - sparse_categorical_accuracy: 0.9425 567/938 [=================>............] - ETA: 0s - loss: 0.4097 - sparse_categorical_accuracy: 0.9428 596/938 [==================>...........] - ETA: 0s - loss: 0.4109 - sparse_categorical_accuracy: 0.9426 625/938 [==================>...........] - ETA: 0s - loss: 0.4157 - sparse_categorical_accuracy: 0.9424 655/938 [===================>..........] - ETA: 0s - loss: 0.4127 - sparse_categorical_accuracy: 0.9423 685/938 [====================>.........] - ETA: 0s - loss: 0.4123 - sparse_categorical_accuracy: 0.9424 715/938 [=====================>........] - ETA: 0s - loss: 0.4091 - sparse_categorical_accuracy: 0.9427 745/938 [======================>.......] - ETA: 0s - loss: 0.4108 - sparse_categorical_accuracy: 0.9427 775/938 [=======================>......] - ETA: 0s - loss: 0.4109 - sparse_categorical_accuracy: 0.9427 805/938 [========================>.....] - ETA: 0s - loss: 0.4152 - sparse_categorical_accuracy: 0.9424 836/938 [=========================>....] - ETA: 0s - loss: 0.4143 - sparse_categorical_accuracy: 0.9426 866/938 [==========================>...] - ETA: 0s - loss: 0.4094 - sparse_categorical_accuracy: 0.9429 896/938 [===========================>..] - ETA: 0s - loss: 0.4041 - sparse_categorical_accuracy: 0.9434 925/938 [============================>.] - ETA: 0s - loss: 0.3990 - sparse_categorical_accuracy: 0.9442 938/938 [==============================] - 2s 2ms/step - loss: 0.3991 - sparse_categorical_accuracy: 0.9444 Epoch 4/10 1/938 [..............................] - ETA: 0s - loss: 0.1693 - sparse_categorical_accuracy: 0.9844 30/938 [..............................] - ETA: 1s - loss: 0.4642 - sparse_categorical_accuracy: 0.9458 60/938 [>.............................] - ETA: 1s - loss: 0.3759 - sparse_categorical_accuracy: 0.9536 90/938 [=>............................] - ETA: 1s - loss: 0.3245 - sparse_categorical_accuracy: 0.9561 120/938 [==>...........................] - ETA: 1s - loss: 0.3519 - sparse_categorical_accuracy: 0.9547 150/938 [===>..........................] - ETA: 1s - loss: 0.3791 - sparse_categorical_accuracy: 0.9504 179/938 [====>.........................] - ETA: 1s - loss: 0.3602 - sparse_categorical_accuracy: 0.9499 209/938 [=====>........................] - ETA: 1s - loss: 0.3601 - sparse_categorical_accuracy: 0.9501 239/938 [======>.......................] - ETA: 1s - loss: 0.3597 - sparse_categorical_accuracy: 0.9499 269/938 [=======>......................] - ETA: 1s - loss: 0.3554 - sparse_categorical_accuracy: 0.9496 299/938 [========>.....................] - ETA: 1s - loss: 0.3517 - sparse_categorical_accuracy: 0.9501 329/938 [=========>....................] - ETA: 1s - loss: 0.3520 - sparse_categorical_accuracy: 0.9495 359/938 [==========>...................] - ETA: 0s - loss: 0.3522 - sparse_categorical_accuracy: 0.9496 388/938 [===========>..................] - ETA: 0s - loss: 0.3510 - sparse_categorical_accuracy: 0.9501 417/938 [============>.................] - ETA: 0s - loss: 0.3466 - sparse_categorical_accuracy: 0.9504 447/938 [=============>................] - ETA: 0s - loss: 0.3430 - sparse_categorical_accuracy: 0.9505 477/938 [==============>...............] - ETA: 0s - loss: 0.3360 - sparse_categorical_accuracy: 0.9507 506/938 [===============>..............] - ETA: 0s - loss: 0.3413 - sparse_categorical_accuracy: 0.9508 535/938 [================>.............] - ETA: 0s - loss: 0.3375 - sparse_categorical_accuracy: 0.9513 564/938 [=================>............] - ETA: 0s - loss: 0.3329 - sparse_categorical_accuracy: 0.9517 594/938 [=================>............] - ETA: 0s - loss: 0.3340 - sparse_categorical_accuracy: 0.9512 623/938 [==================>...........] - ETA: 0s - loss: 0.3361 - sparse_categorical_accuracy: 0.9514 652/938 [===================>..........] - ETA: 0s - loss: 0.3341 - sparse_categorical_accuracy: 0.9516 682/938 [====================>.........] - ETA: 0s - loss: 0.3317 - sparse_categorical_accuracy: 0.9520 711/938 [=====================>........] - ETA: 0s - loss: 0.3302 - sparse_categorical_accuracy: 0.9522 740/938 [======================>.......] - ETA: 0s - loss: 0.3304 - sparse_categorical_accuracy: 0.9520 769/938 [=======================>......] - ETA: 0s - loss: 0.3286 - sparse_categorical_accuracy: 0.9520 799/938 [========================>.....] - ETA: 0s - loss: 0.3291 - sparse_categorical_accuracy: 0.9520 828/938 [=========================>....] - ETA: 0s - loss: 0.3298 - sparse_categorical_accuracy: 0.9520 857/938 [==========================>...] - ETA: 0s - loss: 0.3271 - sparse_categorical_accuracy: 0.9522 887/938 [===========================>..] - ETA: 0s - loss: 0.3246 - sparse_categorical_accuracy: 0.9524 917/938 [============================>.] - ETA: 0s - loss: 0.3189 - sparse_categorical_accuracy: 0.9531 938/938 [==============================] - 2s 2ms/step - loss: 0.3187 - sparse_categorical_accuracy: 0.9535 Epoch 5/10 1/938 [..............................] - ETA: 0s - loss: 0.0664 - sparse_categorical_accuracy: 0.9531 31/938 [..............................] - ETA: 1s - loss: 0.3681 - sparse_categorical_accuracy: 0.9486 61/938 [>.............................] - ETA: 1s - loss: 0.3003 - sparse_categorical_accuracy: 0.9600 90/938 [=>............................] - ETA: 1s - loss: 0.2971 - sparse_categorical_accuracy: 0.9608 119/938 [==>...........................] - ETA: 1s - loss: 0.3057 - sparse_categorical_accuracy: 0.9597 149/938 [===>..........................] - ETA: 1s - loss: 0.3246 - sparse_categorical_accuracy: 0.9576 179/938 [====>.........................] - ETA: 1s - loss: 0.3089 - sparse_categorical_accuracy: 0.9573 208/938 [=====>........................] - ETA: 1s - loss: 0.3040 - sparse_categorical_accuracy: 0.9575 238/938 [======>.......................] - ETA: 1s - loss: 0.3201 - sparse_categorical_accuracy: 0.9566 267/938 [=======>......................] - ETA: 1s - loss: 0.3131 - sparse_categorical_accuracy: 0.9565 297/938 [========>.....................] - ETA: 1s - loss: 0.3039 - sparse_categorical_accuracy: 0.9575 326/938 [=========>....................] - ETA: 1s - loss: 0.3081 - sparse_categorical_accuracy: 0.9577 356/938 [==========>...................] - ETA: 1s - loss: 0.3060 - sparse_categorical_accuracy: 0.9577 385/938 [===========>..................] - ETA: 0s - loss: 0.3061 - sparse_categorical_accuracy: 0.9581 415/938 [============>.................] - ETA: 0s - loss: 0.3092 - sparse_categorical_accuracy: 0.9584 446/938 [=============>................] - ETA: 0s - loss: 0.3130 - sparse_categorical_accuracy: 0.9583 476/938 [==============>...............] - ETA: 0s - loss: 0.3082 - sparse_categorical_accuracy: 0.9586 506/938 [===============>..............] - ETA: 0s - loss: 0.3112 - sparse_categorical_accuracy: 0.9582 535/938 [================>.............] - ETA: 0s - loss: 0.3101 - sparse_categorical_accuracy: 0.9582 565/938 [=================>............] - ETA: 0s - loss: 0.3060 - sparse_categorical_accuracy: 0.9585 595/938 [==================>...........] - ETA: 0s - loss: 0.3045 - sparse_categorical_accuracy: 0.9579 624/938 [==================>...........] - ETA: 0s - loss: 0.3041 - sparse_categorical_accuracy: 0.9578 655/938 [===================>..........] - ETA: 0s - loss: 0.3028 - sparse_categorical_accuracy: 0.9578 685/938 [====================>.........] - ETA: 0s - loss: 0.3028 - sparse_categorical_accuracy: 0.9578 715/938 [=====================>........] - ETA: 0s - loss: 0.3000 - sparse_categorical_accuracy: 0.9580 745/938 [======================>.......] - ETA: 0s - loss: 0.2999 - sparse_categorical_accuracy: 0.9577 775/938 [=======================>......] - ETA: 0s - loss: 0.2965 - sparse_categorical_accuracy: 0.9579 804/938 [========================>.....] - ETA: 0s - loss: 0.2990 - sparse_categorical_accuracy: 0.9577 833/938 [=========================>....] - ETA: 0s - loss: 0.2988 - sparse_categorical_accuracy: 0.9577 863/938 [==========================>...] - ETA: 0s - loss: 0.2954 - sparse_categorical_accuracy: 0.9580 892/938 [===========================>..] - ETA: 0s - loss: 0.2918 - sparse_categorical_accuracy: 0.9582 921/938 [============================>.] - ETA: 0s - loss: 0.2883 - sparse_categorical_accuracy: 0.9585 938/938 [==============================] - 2s 2ms/step - loss: 0.2875 - sparse_categorical_accuracy: 0.9588 Epoch 6/10 1/938 [..............................] - ETA: 0s - loss: 0.2306 - sparse_categorical_accuracy: 0.9844 30/938 [..............................] - ETA: 1s - loss: 0.3352 - sparse_categorical_accuracy: 0.9557 59/938 [>.............................] - ETA: 1s - loss: 0.2677 - sparse_categorical_accuracy: 0.9629 89/938 [=>............................] - ETA: 1s - loss: 0.2472 - sparse_categorical_accuracy: 0.9647 119/938 [==>...........................] - ETA: 1s - loss: 0.2460 - sparse_categorical_accuracy: 0.9639 149/938 [===>..........................] - ETA: 1s - loss: 0.2809 - sparse_categorical_accuracy: 0.9613 179/938 [====>.........................] - ETA: 1s - loss: 0.2679 - sparse_categorical_accuracy: 0.9612 210/938 [=====>........................] - ETA: 1s - loss: 0.2719 - sparse_categorical_accuracy: 0.9608 239/938 [======>.......................] - ETA: 1s - loss: 0.2715 - sparse_categorical_accuracy: 0.9604 268/938 [=======>......................] - ETA: 1s - loss: 0.2695 - sparse_categorical_accuracy: 0.9602 297/938 [========>.....................] - ETA: 1s - loss: 0.2583 - sparse_categorical_accuracy: 0.9614 326/938 [=========>....................] - ETA: 1s - loss: 0.2693 - sparse_categorical_accuracy: 0.9609 357/938 [==========>...................] - ETA: 0s - loss: 0.2674 - sparse_categorical_accuracy: 0.9610 386/938 [===========>..................] - ETA: 0s - loss: 0.2661 - sparse_categorical_accuracy: 0.9609 417/938 [============>.................] - ETA: 0s - loss: 0.2692 - sparse_categorical_accuracy: 0.9609 446/938 [=============>................] - ETA: 0s - loss: 0.2717 - sparse_categorical_accuracy: 0.9609 476/938 [==============>...............] - ETA: 0s - loss: 0.2702 - sparse_categorical_accuracy: 0.9611 506/938 [===============>..............] - ETA: 0s - loss: 0.2740 - sparse_categorical_accuracy: 0.9608 536/938 [================>.............] - ETA: 0s - loss: 0.2772 - sparse_categorical_accuracy: 0.9612 564/938 [=================>............] - ETA: 0s - loss: 0.2769 - sparse_categorical_accuracy: 0.9611 591/938 [=================>............] - ETA: 0s - loss: 0.2781 - sparse_categorical_accuracy: 0.9610 618/938 [==================>...........] - ETA: 0s - loss: 0.2783 - sparse_categorical_accuracy: 0.9610 645/938 [===================>..........] - ETA: 0s - loss: 0.2745 - sparse_categorical_accuracy: 0.9613 672/938 [====================>.........] - ETA: 0s - loss: 0.2752 - sparse_categorical_accuracy: 0.9613 700/938 [=====================>........] - ETA: 0s - loss: 0.2757 - sparse_categorical_accuracy: 0.9616 727/938 [======================>.......] - ETA: 0s - loss: 0.2745 - sparse_categorical_accuracy: 0.9616 754/938 [=======================>......] - ETA: 0s - loss: 0.2734 - sparse_categorical_accuracy: 0.9618 782/938 [========================>.....] - ETA: 0s - loss: 0.2750 - sparse_categorical_accuracy: 0.9619 809/938 [========================>.....] - ETA: 0s - loss: 0.2774 - sparse_categorical_accuracy: 0.9618 837/938 [=========================>....] - ETA: 0s - loss: 0.2774 - sparse_categorical_accuracy: 0.9618 864/938 [==========================>...] - ETA: 0s - loss: 0.2757 - sparse_categorical_accuracy: 0.9618 892/938 [===========================>..] - ETA: 0s - loss: 0.2725 - sparse_categorical_accuracy: 0.9622 919/938 [============================>.] - ETA: 0s - loss: 0.2685 - sparse_categorical_accuracy: 0.9626 938/938 [==============================] - 2s 2ms/step - loss: 0.2688 - sparse_categorical_accuracy: 0.9628 Epoch 7/10 1/938 [..............................] - ETA: 0s - loss: 0.5132 - sparse_categorical_accuracy: 0.9844 29/938 [..............................] - ETA: 1s - loss: 0.2985 - sparse_categorical_accuracy: 0.9655 59/938 [>.............................] - ETA: 1s - loss: 0.2218 - sparse_categorical_accuracy: 0.9735 89/938 [=>............................] - ETA: 1s - loss: 0.2156 - sparse_categorical_accuracy: 0.9728 120/938 [==>...........................] - ETA: 1s - loss: 0.2077 - sparse_categorical_accuracy: 0.9717 149/938 [===>..........................] - ETA: 1s - loss: 0.2496 - sparse_categorical_accuracy: 0.9688 179/938 [====>.........................] - ETA: 1s - loss: 0.2412 - sparse_categorical_accuracy: 0.9688 209/938 [=====>........................] - ETA: 1s - loss: 0.2463 - sparse_categorical_accuracy: 0.9683 239/938 [======>.......................] - ETA: 1s - loss: 0.2470 - sparse_categorical_accuracy: 0.9677 269/938 [=======>......................] - ETA: 1s - loss: 0.2409 - sparse_categorical_accuracy: 0.9671 299/938 [========>.....................] - ETA: 1s - loss: 0.2385 - sparse_categorical_accuracy: 0.9673 330/938 [=========>....................] - ETA: 1s - loss: 0.2449 - sparse_categorical_accuracy: 0.9670 360/938 [==========>...................] - ETA: 0s - loss: 0.2441 - sparse_categorical_accuracy: 0.9668 390/938 [===========>..................] - ETA: 0s - loss: 0.2460 - sparse_categorical_accuracy: 0.9670 420/938 [============>.................] - ETA: 0s - loss: 0.2518 - sparse_categorical_accuracy: 0.9667 449/938 [=============>................] - ETA: 0s - loss: 0.2483 - sparse_categorical_accuracy: 0.9668 479/938 [==============>...............] - ETA: 0s - loss: 0.2434 - sparse_categorical_accuracy: 0.9672 508/938 [===============>..............] - ETA: 0s - loss: 0.2457 - sparse_categorical_accuracy: 0.9668 538/938 [================>.............] - ETA: 0s - loss: 0.2433 - sparse_categorical_accuracy: 0.9671 567/938 [=================>............] - ETA: 0s - loss: 0.2400 - sparse_categorical_accuracy: 0.9675 596/938 [==================>...........] - ETA: 0s - loss: 0.2450 - sparse_categorical_accuracy: 0.9669 625/938 [==================>...........] - ETA: 0s - loss: 0.2469 - sparse_categorical_accuracy: 0.9666 656/938 [===================>..........] - ETA: 0s - loss: 0.2462 - sparse_categorical_accuracy: 0.9664 687/938 [====================>.........] - ETA: 0s - loss: 0.2451 - sparse_categorical_accuracy: 0.9666 717/938 [=====================>........] - ETA: 0s - loss: 0.2427 - sparse_categorical_accuracy: 0.9667 746/938 [======================>.......] - ETA: 0s - loss: 0.2398 - sparse_categorical_accuracy: 0.9668 775/938 [=======================>......] - ETA: 0s - loss: 0.2425 - sparse_categorical_accuracy: 0.9666 805/938 [========================>.....] - ETA: 0s - loss: 0.2437 - sparse_categorical_accuracy: 0.9666 834/938 [=========================>....] - ETA: 0s - loss: 0.2434 - sparse_categorical_accuracy: 0.9667 864/938 [==========================>...] - ETA: 0s - loss: 0.2426 - sparse_categorical_accuracy: 0.9669 893/938 [===========================>..] - ETA: 0s - loss: 0.2408 - sparse_categorical_accuracy: 0.9669 923/938 [============================>.] - ETA: 0s - loss: 0.2370 - sparse_categorical_accuracy: 0.9672 938/938 [==============================] - 2s 2ms/step - loss: 0.2371 - sparse_categorical_accuracy: 0.9674 Epoch 8/10 1/938 [..............................] - ETA: 0s - loss: 0.0166 - sparse_categorical_accuracy: 0.9844 32/938 [>.............................] - ETA: 1s - loss: 0.2881 - sparse_categorical_accuracy: 0.9639 61/938 [>.............................] - ETA: 1s - loss: 0.2152 - sparse_categorical_accuracy: 0.9713 91/938 [=>............................] - ETA: 1s - loss: 0.2326 - sparse_categorical_accuracy: 0.9701 120/938 [==>...........................] - ETA: 1s - loss: 0.2243 - sparse_categorical_accuracy: 0.9707 151/938 [===>..........................] - ETA: 1s - loss: 0.2574 - sparse_categorical_accuracy: 0.9690 181/938 [====>.........................] - ETA: 1s - loss: 0.2473 - sparse_categorical_accuracy: 0.9692 211/938 [=====>........................] - ETA: 1s - loss: 0.2415 - sparse_categorical_accuracy: 0.9691 240/938 [======>.......................] - ETA: 1s - loss: 0.2359 - sparse_categorical_accuracy: 0.9690 270/938 [=======>......................] - ETA: 1s - loss: 0.2355 - sparse_categorical_accuracy: 0.9682 300/938 [========>.....................] - ETA: 1s - loss: 0.2298 - sparse_categorical_accuracy: 0.9687 330/938 [=========>....................] - ETA: 1s - loss: 0.2299 - sparse_categorical_accuracy: 0.9687 360/938 [==========>...................] - ETA: 0s - loss: 0.2267 - sparse_categorical_accuracy: 0.9691 391/938 [===========>..................] - ETA: 0s - loss: 0.2322 - sparse_categorical_accuracy: 0.9692 421/938 [============>.................] - ETA: 0s - loss: 0.2372 - sparse_categorical_accuracy: 0.9689 450/938 [=============>................] - ETA: 0s - loss: 0.2382 - sparse_categorical_accuracy: 0.9692 479/938 [==============>...............] - ETA: 0s - loss: 0.2361 - sparse_categorical_accuracy: 0.9693 509/938 [===============>..............] - ETA: 0s - loss: 0.2361 - sparse_categorical_accuracy: 0.9690 538/938 [================>.............] - ETA: 0s - loss: 0.2340 - sparse_categorical_accuracy: 0.9694 567/938 [=================>............] - ETA: 0s - loss: 0.2298 - sparse_categorical_accuracy: 0.9698 597/938 [==================>...........] - ETA: 0s - loss: 0.2315 - sparse_categorical_accuracy: 0.9695 627/938 [===================>..........] - ETA: 0s - loss: 0.2324 - sparse_categorical_accuracy: 0.9695 657/938 [====================>.........] - ETA: 0s - loss: 0.2334 - sparse_categorical_accuracy: 0.9696 687/938 [====================>.........] - ETA: 0s - loss: 0.2331 - sparse_categorical_accuracy: 0.9695 717/938 [=====================>........] - ETA: 0s - loss: 0.2327 - sparse_categorical_accuracy: 0.9695 747/938 [======================>.......] - ETA: 0s - loss: 0.2343 - sparse_categorical_accuracy: 0.9695 777/938 [=======================>......] - ETA: 0s - loss: 0.2335 - sparse_categorical_accuracy: 0.9695 807/938 [========================>.....] - ETA: 0s - loss: 0.2352 - sparse_categorical_accuracy: 0.9693 837/938 [=========================>....] - ETA: 0s - loss: 0.2359 - sparse_categorical_accuracy: 0.9693 868/938 [==========================>...] - ETA: 0s - loss: 0.2338 - sparse_categorical_accuracy: 0.9694 899/938 [===========================>..] - ETA: 0s - loss: 0.2324 - sparse_categorical_accuracy: 0.9695 929/938 [============================>.] - ETA: 0s - loss: 0.2292 - sparse_categorical_accuracy: 0.9699 938/938 [==============================] - 2s 2ms/step - loss: 0.2286 - sparse_categorical_accuracy: 0.9700 Epoch 9/10 1/938 [..............................] - ETA: 0s - loss: 0.0915 - sparse_categorical_accuracy: 0.9844 30/938 [..............................] - ETA: 1s - loss: 0.2903 - sparse_categorical_accuracy: 0.9677 60/938 [>.............................] - ETA: 1s - loss: 0.2410 - sparse_categorical_accuracy: 0.9729 90/938 [=>............................] - ETA: 1s - loss: 0.2310 - sparse_categorical_accuracy: 0.9731 120/938 [==>...........................] - ETA: 1s - loss: 0.2317 - sparse_categorical_accuracy: 0.9729 151/938 [===>..........................] - ETA: 1s - loss: 0.2429 - sparse_categorical_accuracy: 0.9716 181/938 [====>.........................] - ETA: 1s - loss: 0.2329 - sparse_categorical_accuracy: 0.9719 212/938 [=====>........................] - ETA: 1s - loss: 0.2240 - sparse_categorical_accuracy: 0.9721 242/938 [======>.......................] - ETA: 1s - loss: 0.2228 - sparse_categorical_accuracy: 0.9724 273/938 [=======>......................] - ETA: 1s - loss: 0.2136 - sparse_categorical_accuracy: 0.9725 304/938 [========>.....................] - ETA: 1s - loss: 0.2152 - sparse_categorical_accuracy: 0.9725 334/938 [=========>....................] - ETA: 1s - loss: 0.2119 - sparse_categorical_accuracy: 0.9726 361/938 [==========>...................] - ETA: 0s - loss: 0.2128 - sparse_categorical_accuracy: 0.9725 389/938 [===========>..................] - ETA: 0s - loss: 0.2210 - sparse_categorical_accuracy: 0.9720 419/938 [============>.................] - ETA: 0s - loss: 0.2213 - sparse_categorical_accuracy: 0.9718 448/938 [=============>................] - ETA: 0s - loss: 0.2252 - sparse_categorical_accuracy: 0.9717 479/938 [==============>...............] - ETA: 0s - loss: 0.2226 - sparse_categorical_accuracy: 0.9717 508/938 [===============>..............] - ETA: 0s - loss: 0.2218 - sparse_categorical_accuracy: 0.9714 538/938 [================>.............] - ETA: 0s - loss: 0.2202 - sparse_categorical_accuracy: 0.9718 568/938 [=================>............] - ETA: 0s - loss: 0.2180 - sparse_categorical_accuracy: 0.9722 598/938 [==================>...........] - ETA: 0s - loss: 0.2177 - sparse_categorical_accuracy: 0.9719 628/938 [===================>..........] - ETA: 0s - loss: 0.2196 - sparse_categorical_accuracy: 0.9717 659/938 [====================>.........] - ETA: 0s - loss: 0.2198 - sparse_categorical_accuracy: 0.9717 689/938 [=====================>........] - ETA: 0s - loss: 0.2216 - sparse_categorical_accuracy: 0.9716 719/938 [=====================>........] - ETA: 0s - loss: 0.2203 - sparse_categorical_accuracy: 0.9716 750/938 [======================>.......] - ETA: 0s - loss: 0.2190 - sparse_categorical_accuracy: 0.9715 780/938 [=======================>......] - ETA: 0s - loss: 0.2195 - sparse_categorical_accuracy: 0.9714 810/938 [========================>.....] - ETA: 0s - loss: 0.2185 - sparse_categorical_accuracy: 0.9714 840/938 [=========================>....] - ETA: 0s - loss: 0.2178 - sparse_categorical_accuracy: 0.9714 870/938 [==========================>...] - ETA: 0s - loss: 0.2171 - sparse_categorical_accuracy: 0.9715 901/938 [===========================>..] - ETA: 0s - loss: 0.2156 - sparse_categorical_accuracy: 0.9716 931/938 [============================>.] - ETA: 0s - loss: 0.2150 - sparse_categorical_accuracy: 0.9717 938/938 [==============================] - 2s 2ms/step - loss: 0.2157 - sparse_categorical_accuracy: 0.9718 Epoch 10/10 1/938 [..............................] - ETA: 1s - loss: 0.1469 - sparse_categorical_accuracy: 0.9688 31/938 [..............................] - ETA: 1s - loss: 0.2365 - sparse_categorical_accuracy: 0.9708 61/938 [>.............................] - ETA: 1s - loss: 0.2056 - sparse_categorical_accuracy: 0.9767 91/938 [=>............................] - ETA: 1s - loss: 0.2083 - sparse_categorical_accuracy: 0.9744 121/938 [==>...........................] - ETA: 1s - loss: 0.1929 - sparse_categorical_accuracy: 0.9756 150/938 [===>..........................] - ETA: 1s - loss: 0.2089 - sparse_categorical_accuracy: 0.9739 178/938 [====>.........................] - ETA: 1s - loss: 0.1993 - sparse_categorical_accuracy: 0.9740 205/938 [=====>........................] - ETA: 1s - loss: 0.2007 - sparse_categorical_accuracy: 0.9738 232/938 [======>.......................] - ETA: 1s - loss: 0.2009 - sparse_categorical_accuracy: 0.9735 260/938 [=======>......................] - ETA: 1s - loss: 0.2015 - sparse_categorical_accuracy: 0.9736 287/938 [========>.....................] - ETA: 1s - loss: 0.2001 - sparse_categorical_accuracy: 0.9735 315/938 [=========>....................] - ETA: 1s - loss: 0.2015 - sparse_categorical_accuracy: 0.9738 343/938 [=========>....................] - ETA: 1s - loss: 0.2011 - sparse_categorical_accuracy: 0.9744 371/938 [==========>...................] - ETA: 1s - loss: 0.2001 - sparse_categorical_accuracy: 0.9746 398/938 [===========>..................] - ETA: 0s - loss: 0.2064 - sparse_categorical_accuracy: 0.9743 426/938 [============>.................] - ETA: 0s - loss: 0.2105 - sparse_categorical_accuracy: 0.9741 455/938 [=============>................] - ETA: 0s - loss: 0.2090 - sparse_categorical_accuracy: 0.9742 483/938 [==============>...............] - ETA: 0s - loss: 0.2034 - sparse_categorical_accuracy: 0.9742 510/938 [===============>..............] - ETA: 0s - loss: 0.2043 - sparse_categorical_accuracy: 0.9741 537/938 [================>.............] - ETA: 0s - loss: 0.2069 - sparse_categorical_accuracy: 0.9743 565/938 [=================>............] - ETA: 0s - loss: 0.2089 - sparse_categorical_accuracy: 0.9741 592/938 [=================>............] - ETA: 0s - loss: 0.2089 - sparse_categorical_accuracy: 0.9739 620/938 [==================>...........] - ETA: 0s - loss: 0.2097 - sparse_categorical_accuracy: 0.9736 647/938 [===================>..........] - ETA: 0s - loss: 0.2075 - sparse_categorical_accuracy: 0.9737 674/938 [====================>.........] - ETA: 0s - loss: 0.2112 - sparse_categorical_accuracy: 0.9735 701/938 [=====================>........] - ETA: 0s - loss: 0.2098 - sparse_categorical_accuracy: 0.9737 729/938 [======================>.......] - ETA: 0s - loss: 0.2089 - sparse_categorical_accuracy: 0.9737 756/938 [=======================>......] - ETA: 0s - loss: 0.2119 - sparse_categorical_accuracy: 0.9737 784/938 [========================>.....] - ETA: 0s - loss: 0.2141 - sparse_categorical_accuracy: 0.9736 811/938 [========================>.....] - ETA: 0s - loss: 0.2152 - sparse_categorical_accuracy: 0.9735 838/938 [=========================>....] - ETA: 0s - loss: 0.2142 - sparse_categorical_accuracy: 0.9734 865/938 [==========================>...] - ETA: 0s - loss: 0.2132 - sparse_categorical_accuracy: 0.9734 892/938 [===========================>..] - ETA: 0s - loss: 0.2137 - sparse_categorical_accuracy: 0.9733 919/938 [============================>.] - ETA: 0s - loss: 0.2120 - sparse_categorical_accuracy: 0.9735 938/938 [==============================] - 2s 2ms/step - loss: 0.2115 - sparse_categorical_accuracy: 0.9736 <tensorflow.python.keras.callbacks.History at 0x7f401a9bff98>
model.evaluate(test_dataset)
1/157 [..............................] - ETA: 0s - loss: 0.2801 - sparse_categorical_accuracy: 0.9844 35/157 [=====>........................] - ETA: 0s - loss: 0.9197 - sparse_categorical_accuracy: 0.9415 69/157 [============>.................] - ETA: 0s - loss: 1.0075 - sparse_categorical_accuracy: 0.9386 103/157 [==================>...........] - ETA: 0s - loss: 0.8572 - sparse_categorical_accuracy: 0.9469 136/157 [========================>.....] - ETA: 0s - loss: 0.7464 - sparse_categorical_accuracy: 0.9512 157/157 [==============================] - 0s 2ms/step - loss: 0.7278 - sparse_categorical_accuracy: 0.9533 [0.7278250455856323, 0.9532999992370605]