First of all,
I must use the function fit_generator() , model.fit_generator(...)
In this situation, I have to visualize the input image and output image by tensorboard.
Even though I use the tensorboard callback like below, it didn't visualize image at all.
tensorboard_callback = TensorBoard(log_dir='./logs', write_images=True)
model.fit_generator(..., callbacks=[tensorboard_callback])
I saw many posts related to this issue(a,b) and found some hints to implement custom callback class. However, there was no way to get the input image and output image to write into tensorboard.
Please help, thanks
