Rajendra Choudhary comments

Repositories
Issues
Comments

Results 4 comments of


                                            Rajendra Choudhary

Questions about PyTorch version

> Cannot re-initialize CUDA in forked subprocess. To use CUDA with multiprocessing, you must use the 'spawn' start method To resolve this: Change 'self.num_workers = 4' to 'self.num_workers = 0'...

ValueError: Cannot feed value of shape (64, 218, 178, 3) for Tensor 'real_images:0', which has shape '(64, 64, 64, 3)'

To resolve this: add img_bgr = cv2.resize(img_bgr, (200, 200)) in utils.py after img_bgr = cv2.imread(path)

It did not work when I try to convert the default model "chatglm2" to "llama2"

I am also facing the same issue. I am running on AWS g5.8xlarge Were you able to solve it?

It did not work when I try to convert the default model "chatglm2" to "llama2"

Since LlamaTokenizer does not have value for tokenizer.pad_token_id, its value is None. When this tokenizer is used to calculate the value of 'labels' it gives the error `TypeError: 'NoneType' object...