search

pykt-team / pykt-toolkit

pyKT: A Python Library to Benchmark Deep Learning based Knowledge Tracing Models
https://pykt.org
MIT License
196 stars 53 forks source link

About run environment #53

Closed coder-zc closed 1 year ago

coder-zc commented 1 year ago

Excuse me, can the model run only in the GPU environment that supports cuda?

meta-tabchen commented 1 year ago

Hi~thank you for your interest in our work. Yes, now we only support GPU machines, we will support CPU machines in the future.

And you can try to modify pykt/datasets/data_loader.py where we use from torch.cuda import FloatTensor, LongTensor to set the default machine to GPU, maybe you can change this line to support CPU.

coder-zc commented 1 year ago

Thank you for your reply！i tried to change from torch.cuda import FloatTensor, LongTensor into from torch import FloatTensor, LongTensor, which can run in the CPU.

meta-tabchen commented 1 year ago

Thank you for your results, we will update this to a future version.

coder-zc commented 1 year ago

Hello! i want to konw why generate window_sequences when testing. Could you briefly describe the sequence_window of how and why to do this，which generate a large number of sequences.

meta-tabchen commented 1 year ago

Hello! i want to konw why generate window_sequences when testing. Could you briefly describe the sequence_window of how and why to do this，which generate a large number of sequences.

Hello ! here has some discussions, if that does not answer your question, please let us know.