Dear fellows, I found a problem that I intended to evaluate(i.e., test) the model without running "bash ./download_models.sh". But it turned out to be an error that denoted there were no ".pt" files in the folder. What I only could do to successfully run "evaluate.py" is to download the pretrained models. What I truly mean is that the models(i.e., the ".pt" files) could not be saved. So could you please tell me how to tackle this problem? If the models cannot be saved, then if I wanna train and test on my own datasets, it could be the same error that I could not run "evaluate.py" without no model files being saved.
Thanks a lot!
Dear fellows, I found a problem that I intended to evaluate(i.e., test) the model without running "bash ./download_models.sh". But it turned out to be an error that denoted there were no ".pt" files in the folder. What I only could do to successfully run "evaluate.py" is to download the pretrained models. What I truly mean is that the models(i.e., the ".pt" files) could not be saved. So could you please tell me how to tackle this problem? If the models cannot be saved, then if I wanna train and test on my own datasets, it could be the same error that I could not run "evaluate.py" without no model files being saved. Thanks a lot!