Breaking News: Grepper is joining You.com. Read the official announcement!

tensor in keras

Innocent Iguana answered on April 17, 2024 Popularity 7/10 Helpfulness 5/10

answer tensor in keras

tensor in keras

Comment

# Also make sure to see TENSORFLOW BASICS

# Import Keras backend
import tensorflow.keras.backend as K
# Get the input and output tensors of a model layer
inp = model.layers[0].input
out = model.layers[0].output
# Function that maps layer inputs to outputs
inp_to_out = K.function([inp], [out])
# We pass and input and get the output we'd get in that first layer
print(inp_to_out([X_train.values])) # make sure to convert into numpy arrays

# Visualize change in neuron outputs
import matplotlib.pyplot as plt
def plot():
  fig, ax = plt.subplots()
  plt.scatter(layer_output[:, 0], layer_output[:, 1],c = y_test,edgecolors='none')
  plt.title('Epoch: {}, Test Accuracy: {:3.1f} %'.format(i+1, test_accuracy * 100.0))
  plt.show()

for i in range(0, 21):
  	# Train model for 1 epoch
    h = model.fit(X_train, y_train, batch_size = 16, epochs = 1, verbose = 0)
    if i%4==0: 
      # Get the output of the first layer
      layer_output = inp_to_out([X_test.values])[0]
      # Evaluate model accuracy for this epoch
      test_accuracy = model.evaluate(X_test, y_test)[1] 
      # Plot 1st vs 2nd neuron output for the two class problem
      plot()

Popularity 7/10 Helpfulness 5/10 Language python

Source: Grepper

Tags: keras python tensor

Link to this answer
Share Copy Link

Contributed on Apr 17 2024

Innocent Iguana

0 Answers Avg Quality 2/10

tensor in keras

Contents

More Related Answers

tensor in keras

Grepper

Documentation

Social

Legal

Contact

Oops, You will need to install Grepper and log-in to perform this action.