Monami Banerjee

I am seeing the same error. Apparently the newer version of ONNX does not have these files. Can you please share the torch version used? That will help to narrow...

Updating requirements.txt, as shown below, solved my issue. ``` EasyDict==1.7 opencv-python==3.4.8.29 shapely==1.6.4 Cython scipy pandas pyyaml json_tricks scikit-image yacs>=0.1.5 tensorboardX tensorboard torch torchvision numpy==1.20 ```

The error occurred in [this line](https://github.com/YueLiao/CDN/blob/main/models/cdn.py#L163).

This is helpful. Thank you! ``` pip3 install --upgrade --force-reinstall tokenizers==0.10.3 transformers==4.5.1 timm==0.5.4 pip install --upgrade --force-reinstall torch==1.7.1+cu110 torchvision==0.8.2+cu110 -f https://download.pytorch.org/whl/torch_stable.html export CUBLAS_WORKSPACE_CONFIG=:4096:8 python -m torch.distributed.launch --nproc_per_node=8 --use_env main.py --dataset_config...

Same error with only by changing the model to `glm4v-9b-chat` from `llava1_6-mistral-7b-instruct` in the first DPO example [here](https://github.com/modelscope/ms-swift/blob/main/docs/source_en/Multi-Modal/human-preference-alignment-training-documentation.md#dpo). ``` CUDA_VISIBLE_DEVICES=0 \ swift rlhf \ --rlhf_type dpo \ --model_type glm4v-9b-chat \...

After changing the model, there are three `batch` keys are missing, `prompt_input_ids`, `prompt_pixel_values`, and `prompt_image_sizes`. And now there is an additional key `prompt_images`.

This update with the following command gave the following error: ``` CUDA_VISIBLE_DEVICES=0 \ swift rlhf \ --rlhf_type dpo \ --model_type glm4v-9b-chat \ --beta 0.1 \ --sft_beta 0.1 \ --sft_type lora...

Created the issue **DPO training error UnboundLocalError: local variable 'num_patches' referenced before assignment #1734**.

Hello @Jintao-Huang, Sorry for the delayed response. Actually the above solution did not resolve the issue. The updated error with the above command is, ``` (swift) m.banerjee@PHYVDGPU03PRMV:/VDIL_COREML/m.banerjee/ms-swift$ NPROC_PER_NODE=3 \ CUDA_VISIBLE_DEVICES=0,1,2,3...

Please re-open this issue until resolved.