Datou


这个AI项目真好玩，但对小白来说，运行起来基本就到头了，想要更深入了解这个项目甚至动手改这个项目，太难了。 比如我一开始尝试在test.py里面增加游戏声音，失败了，只能断断续续出点声，都听不出来和游戏有没有关系。 我的显卡是4090，用林哥的训练参数跑不满，我跟chatgpt聊怎么可以最大化训练效率，它告诉我可以调几个参数，调完之后cpu满了，显卡还是比较空闲，再问它，它说有些参数占用cpu，有些参数占用显卡，你自己多试试。对这些参数缺乏基本的认知，瞎蒙的效率就很低。 怎么理解TensorBoard里的曲线和数值，它们理论上可以指导训练参数的设置对吧，看了几个TensorBoard教程，和AI对战项目对不上号。 我还想调整训练的设置，改成每次训练输一场结束或赢两场结束，而不是打完一场就结束，没找到代码位置。 训练之后的结果在哪，如何变成林哥那种ppo_ryu_2500000_steps_updated.zip在test.py里调用，不知道。 代码量不算大，但代码背后的所涉及的知识太多了 T_T  

ERROR - failed to load model from ./models/gpt4all-lora-quantized-ggml.bin

6

## Current Behavior The default model file (gpt4all-lora-quantized-ggml.bin) already exists. Do you want to replace it? Press B to download it with a browser (faster). [Y,N,B]?N Skipping download of model...

gr.Audio() got an unexpected keyword argument 'info'

6

after run python -m openvoice_app --share, i got this: > Loaded checkpoint 'checkpoints/base_speakers/EN/checkpoint.pth' > missing/unexpected keys: [] [] > Loaded checkpoint 'checkpoints/base_speakers/ZH/checkpoint.pth' > missing/unexpected keys: [] [] > Loaded checkpoint...

Can it be installed under windows?

3

When I got to the last step of the installation, I got an error message saying that WSL is required. 

confidence threshold调低才能检测成功，但是调低就会把错误的区域检测进来

3

confidence threshold 0.05  confidence threshold 0.10 

Could you please provide regional crop and paste functions?

1

There is a facial crop and paste function in WAS node suit, which can extract the facial area, redraw it and paste it back to the original image. But it...

how to "Prepares face embeds for the generation"?

1

Can I upload an ordinary portrait and then embed the image through comfyui's workflow?