Sorry, I am getting acquainted with tensorflow, on the manufacturer's website, code for a quick start. Output to the console on the right. If you use network creation for example using tf.nn.conv2d and output

train_accuracy = accuracy.eval(feed_dict={ x: batch[0], y_: batch[1], keep_prob: 1.0}) print('step %d, training accuracy %g' % (i, train_accuracy))

it works fine. The code uses the keras module, the result is unknown characters. TensorFlow-gpu1.5. How can I fix it, do not tell me. Code:

 import tensorflow as tf mnist = tf.keras.datasets.mnist (x_train, y_train),(x_test, y_test) = mnist.load_data() x_train, x_test = x_train / 255.0, x_test / 255.0 model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(input_shape=(28, 28)), tf.keras.layers.Dense(512, activation=tf.nn.relu), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(10, activation=tf.nn.softmax) ]) model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy']) model.fit(x_train, y_train, epochs=5) model.evaluate(x_test, y_test)

right purple output

  • This is a mnist data download, but why is it so incomprehensible. It should not be like this. - Alexandr1234567890
  • If you're talking about displaying text, then you just need to run this script in cmd - Andrio Skur
  • @ Alexandr1234567890, I do not quite understand the essence of the question. If I run the code from your question in iPython / Jupyter, then everything works fine. mnist.load_data() remarkably first downloads the data, and then returns them already broken into the training and test date of the set ... - MaxU

1 answer 1

Found on the Internet for those who will also start with tensorFlow

 import tensorflow as tf import sys import pickle import gzip #mnist = tf.keras.datasets.mnist f = gzip.open('mnist.pkl.gz', 'rb') if sys.version_info < (3,): data = pickle.load.load(f) else: data = pickle.load(f, encoding='bytes') f.close() (x_train, y_train),(x_test, y_test) = data x_train, x_test = x_train / 255.0, x_test / 255.0 model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(input_shape=(28, 28)), tf.keras.layers.Dense(512, activation=tf.nn.relu), tf.keras.layers.Dropout(0.2), tf.keras.layers.Dense(10, activation=tf.nn.softmax) ]) model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy']) model.fit(x_train, y_train, epochs=5) model.fit() model.evaluate(x_test, y_test)

You need to download the archive mnist.pkl.gz and execute this code. thank