A series of Jupyter notebooks that walk you through the fundamentals of Machine Learning and Deep Learning in Python using Scikit-Learn, Keras and TensorFlow 2.
Apache License 2.0
27.26k
stars
12.6k
forks
source link
[QUESTION] Plotting loss with the Deep Convolutional GAN #563
Describe what is unclear to you
When creating autoencoders, we were using fit() to produce the history, which could then be used to plot the loss across the training and validation periods, such as on p. 590:
history = variational_ae.fit(X_train, X_train, epochs=25, batch_size=128,
validation_data=(X_valid, X_valid))
However, when creating the GANs and deep convolutional GANs, we do not use fit(), we use the custom train_gan function:
Describe what is unclear to you When creating autoencoders, we were using fit() to produce the history, which could then be used to plot the loss across the training and validation periods, such as on p. 590:
However, when creating the GANs and deep convolutional GANs, we do not use fit(), we use the custom train_gan function:
If we wanted to plot the loss of both the discriminator and generator across all epochs in the example on page 599, how would we go about this?
Thanks!