About the module train

[xxxmmm00]

Hello！Do you train your model on GPU? During training, I found that the code only runs on the CPU. Now I want to put the code on the GPU. Is there any way?

[sharathadavanne]

If you have a compatible GPU, Keras should automatically train on it.
Check this issue - https://stackoverflow.com/questions/45662253/can-i-run-keras-model-on-gpu

[xxxmmm00]

Thank you very much for your help

[xxxmmm00]

Sorry, now the code can run on GPU, but the computer memory occupation rate increases gradually with the code running until it crashes. I can't solve this problem. Is there any way to effectively solve this problem?

[muuda]

Sorry, now the code can run on GPU, but the computer memory occupation rate increases gradually with the code running until it crashes. I can't solve this problem. Is there any way to effectively solve this problem?

Hello, can you tell me how to train with gpu? I have been groping for a day without solving this problem. My Python version is 3.7 and cuda version is 11.2. What version should I use for tensorflow and keras? Is an updated version required? Do I need to change the code after updating the version? Thank you very much for your help！！！

[xxxmmm00]

Because it has been a long time, I can't remember clearly. You need to download the GPU version of TensorFlow and the corresponding cudatoolkit version, and the model code can be trained on the GPU without making changes. The code definitely needs to be changed. If you have configured a higher version of TensorFlow, corresponding solutions to these issues can be found in other communities.

[muuda]

Because it has been a long time, I can't remember clearly. You need to download the GPU version of TensorFlow and the corresponding cudatoolkit version, and the model code can be trained on the GPU without making changes. The code definitely needs to be changed. If you have configured a higher version of TensorFlow, corresponding solutions to these issues can be found in other communities.

Thank you! I will try it

[muuda]

Sorry, now the code can run on GPU, but the computer memory occupation rate increases gradually with the code running until it crashes. I can't solve this problem. Is there any way to effectively solve this problem?

Sorry to bother you again. I also encountered this problem. How did you solve it at that time? Thank you very much for your help.