Open agussa opened 4 years ago
Same question for citation.
@agussa I would like to ask do you know how to retrain the network? How can I read all the training data as x, and label as y?
if you have enough RAM/ GPU memory than you can do this
How can I read all the training data as x, and label as y?
But, if you don't have enough RAM/GPU RAM, then it better to use keras.sequence as generator or tf.keras.sequence to fetch one by one and feed to network, or per-batch as you specify it.
Do you have any code samples? I am not quite familiar with the generator... Can I use the function create_data in the notebook to read it?
This is not a problem. Just a question. your work is great! It inspired me to modify to suit my problem. How to cite your work ?
Thanks.