CosmicBanana777
commented
1 year ago Been tinkering with this over the past few days, I've found that the last for loop in main.py used for creating the output string wasn't working correctly with my model. I've edited the for loop extensively and have gotten it working. I think it may be due to how my model generates things, based on how I've formatted the data from Discord Chat Exporter.
In addition, switching to the proper Pytorch CUDA version that matches mine (11.8) worked (had to find it under the preview builds), although I get userwarnings that ShardedTensor is being depreciated for Dtensor. It doesn't seem to affect anything though and I don't feel like reverting to older versions of CUDA
Hi, I trained a pytorch_model.bin using google colab and put it and the config.json within the trained_model folder in the gravital folder. However, when I try to do --test, it only shows the generation config and nothing is output. A little new to this, any help is appreciated.
In addition, after installing pytorch, I keep getting this warning:
UserWarning: Failed to load image Python extension: Could not find module 'C:\Users\...\AppData\Local\Programs\Python\Python310\Lib\site-packages\torchvision\image.pyd' (or one of its dependencies). Try using the full path with constructor syntax. warn(f"Failed to load image Python extension: {e}")Anyone know what's causing this?