GY_Zhang comments

Results 4 comments of


                                            GY_Zhang

Some questions about the eval metrics of this paper.

Well, now I know swin-t == swin-transformer. But the last two question still confuse me a lot now.

Some questions about the eval metrics of this paper.

And in my environment, my result is as this picture: ![image](https://github.com/chou141253/FGVC-PIM/assets/48321774/d447848d-355b-43eb-be03-3c48310dedf6) So it confuses me that `highest` results is different with `layer` 1-4 and the `combiner` layer.

RuntimeError: CUDA error: invalid device ordinal CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

Thank you for your timely reply. I have used single GPU for training and successfully, I will try your new code right now.

There are some bugs in eval.py

After turnning "rag" to "memorag", I find another question ![image](https://github.com/user-attachments/assets/bbc88f4e-1d7c-4c34-a4e3-6342276b993f) ![image](https://github.com/user-attachments/assets/43a5ab80-195c-43a1-bc9d-8bf1e2578320) Why the value of query is x["context"][0] and the value of context is x["question"][0] ?