black_cat

In my thougnt,the in_channel and out_channel should be same in UNet, but in x4-upscaling.yaml I think this use the latent low_scale_noise (3 channels) and the img(4 channels),so it has 7...

I run the gradio_app.py to use, the same input image is used, but much slower than huggingface webui. I notice that GPU is not used, torch.cuda.is_available() is True ,its there...

如何构建prompt和history 能够让知识库检索回答按照选择题返回

1

**问题描述 / Problem Description** 用简洁明了的语言描述这个问题 / Describe the problem in a clear and concise manner. **复现问题的步骤 / Steps to Reproduce** 1. 执行 '...' / Run '...' 2. 点击 '...' /...

知识库检索的代码

2

我想修改一下知识库检索的代码 因为目前的召回率比较差 想改为混合检索的方式 请问在哪个位置 好像0.1版本的检索是根据上下文的，请问一下代码位置 想做参考 还有注意到0.3.x版本有改进这个检索模式的想法 问一下什么时候能上线呢

Why I get killed?

8

I use CLI `python main.py --train --base configs/stableSRNew/v2-finetune_text_T_512.yaml --gpus GPU_ID, --name NAME --scale_lr False` and i get killed  At first ,i reduce the batch_size and num_workers...

How can I visualize the graph and relationship?

23

Thank you for your great work！ I have run the whole process, but it is hard to see the graph clearly. The graph data are stored in the .parquet files,...

[Bug]: Qwen2 compete error

1

### Bug Description I'm try to use Qwen2 model by ollama , but meet timeout error by using llm.compete, timed out occured. but when I use llm.chat,it can return reply...