Note: Performing the SVD in a CUDA environment will accelerate the process; however, due to an unknown error, we conducted it using the CPU. Other details follow SVD-GCN.
Python==3.9 | PyTorch==1.12.1 | freerec==0.4.3
We provide configs and experimental logs. You can re-run them and try some other hyperparameters:
python main.py --config=configs/xxx.yaml