Chaofeng Chen

请参考示例代码`inference_iqa.py`，以及测试代码: ``` # example for FR metric with dirs python inference_iqa.py -m [metric_name_in_lower_case] -i ./ResultsCalibra/dist_dir[dist_img] -r ./ResultsCalibra/ref_dir[ref_img] # example for NR metric with single image python inference_iqa.py -m [metric_name_in_lower_case] -i...

Thanks for suggestion. It is not possible to give range or better value indications at this repo, especially for recent DL methods. Because for learning based approaches, it depends on...

Thanks to your feedback. Training with `DataParallel` can sometimes be annoying. It is better to use `DistributedDataParallel` as recommended by [official document](https://pytorch.org/docs/stable/generated/torch.nn.DataParallel.html) You may `git pull` to the latest version...

Here is the debug message from ChatGPT. It is likely because the pytorch is not corrected installed with CUDNN. > It seems you've uploaded an image of an error message...

Thanks to your interest. The model now supports: - CompVis/stable-diffusion-v1-4 - runwayml/stable-diffusion-v1-5 - stabilityai/stable-diffusion-2-1-base You may use them with corresponding `model_id` and lora weight paths in demo code.

Sorry, but sdxl is not something we can handle right now. Besides, it's way too slow compared to the turbo version. We're working hard to find a better reward model...

The test codes and weight have already been released. Please refer to the colab demo: https://colab.research.google.com/drive/1pC6lzFR4hohwWyBgnkeZUghi2jqjl9gI?usp=sharing

I appreciate your enthusiasm and curiosity about my work. I am currently working on improving the code and the model, aiming for more robust results. This is a challenging and...

Thank you for your patience and understanding. While the current performance may not fully meet our expectations, we acknowledge that there has been a significant delay in releasing the code....

We appreciate your enthusiasm and interest in trying out our method. However, it's important to note that **our paper primarily serves to validate the feasibility of the technical approach.** Currently,...