Answer the question
In order to leave comments, you need to log in
How to feed a picture into a neural network that the network itself generated for the first time?
I have a mesh convolutional vae. I want to experiment and submit several times the picture that the network itself issued. But I don't understand how to take the resulting image and submit it already to the input. I have an option to use predict. But I'm sure you can somehow do it through epoch , but I don't understand how.
Answer the question
In order to leave comments, you need to log in
Can I find out where the confidence came from that this can be done through epoch?
When we train the network to display the process of learning and changing pictures, we do this through a function from epochs
# Функция, которую будем запускать после каждой эпохи
def on_epoch_end(epoch, logs):
if epoch in save_epochs:
clear_output()
# Сравнение реальных и декодированных цифр
decoded = vae.predict()
plot_img(#функция для выведения изображений)</code
типа такого
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question