search

maxjcohen / transformer

Implementation of Transformer model (originally from Attention is All You Need) applied to Time Series.
https://timeseriestransformer.readthedocs.io/en/latest/
GNU General Public License v3.0
842 stars 165 forks source link

Issue while training the model #50

Closed sarraAyed closed 2 years ago

sarraAyed commented 2 years ago

First, I would like to thank you for creating and publishing this repo.
Also, thank you for taking the time to answer poeple's questions I was trying to run the train part and I had this problem RuntimeError: CUDA out of memory. Tried to allocate 1.72 GiB (GPU 0; 2.00 GiB total capacity; 109.01 MiB already allocated; 1.14 GiB free; 112.00 MiB reserved in total by PyTorch) Is there another way for me to train the model ? Thank you so much for you attention

maxjcohen commented 2 years ago

It seems you do not have enough memory in your GPU. You could try to run the model on your CPU, which would be slower, but allow computation on the RAM.