SiWeiTu

Results 18 issues of SiWeiTu

I used the code provided in the tutorial but only generated the first two Chinese characters of the audio why?

can you renew the connection of lama?

When reconstructing a video, num_ Frames only supports 16, and the duration of combining reconstructed images into videos is too short. Is there any solution

![image](https://github.com/jxhe/unify-parameter-efficient-tuning/assets/67264710/ec02aa3d-377e-46fd-b122-e65f6bc78d59) transformers.adapter only have prefix_tuning and adapter,where is the lora

Hello, thank you for your kind courses. but where can I find the Chinese version of the courses? Thank you.

### Describe the issue Thank you for your work. Recently, I've been looking at the pretraining code of llava v1. However, I noticed that the `mm_vision_tower` parameter is not present...

I tried running Python 3.8 due to other project requirements, but it failed. Has anyone run successfully in the 3.8 environment? Thank you

what can i do to change the background color to black with the detected object?like the example as follow ![Snipaste_2023-10-19_22-53-58](https://github.com/CASIA-IVA-Lab/FastSAM/assets/67264710/95315c57-040f-476a-a4ec-1b7f57c4703e)

### Feature request When I use the generate() function of a language model for inference, the kv-cache is also stored in the GPU memory. Is there any way to clear...

Feature request
Generation
Cache