-
使用 JIT 导出 flow 时,出现 字典参数展开 **argv,无法导出
请问是否有替代 matcha flow 的方案,能支持 JIT 导出
排查发现是 flow.decoder 部分,调用 third_party/Matcha-TTS/matcha/models/components/transformer.py
![image](https://github.com…
-
python infer_video.py -m gmflow+pervfi-vb -data input -fps 11
Testing on Dataset: input
Running VFI method : gmflow+pervfi-vb
TMP (temporary) Dir: /tmp/tmp62froo40
VIS (visualize) Dir: out…
-
**Describe the bug**
In addition to generating onnx models for the standard LLM, we also want to be able to create 2-layer models, or otherwise abridged models for faster proof of concept. Optimally…
-
## Description
There are many wallet models exposing data in various ways. Usually they are later transformed on UI side before being finally consumed by UI views. It would be beneficial to have th…
-
python3 webui.py --port 7865 --model_dir pretrained_models/CosyVoice-300M
2024-10-21 12:16:38,971 - modelscope - INFO - PyTorch version 2.0.1 Found.
2024-10-21 12:16:38,971 - modelscope - INFO - Loa…
-
Crashes when attempting to construct a `codeable_model`, OK otherwise:
```
Traceback (most recent call last):
File "/Users/abakhtin22/Documents/TempNet/code2DFD/code2DFD.py", line 115, in
…
-
### Validations
- [X] I believe this is a way to improve. I'll try to join the [Continue Discord](https://discord.gg/NWtdYexhMs) for questions
- [X] I'm not able to find an [open issue](https://githu…
-
在使用预置的pretrained model时,包括 pretrained_models/CosyVoice-300M-Instruct 和 pretrained_models/CosyVoice-300M,都会报同一个错误:
ValueError: The provided filename pretrained_models/CosyVoice-300M/flow.encoder.fp32.…
-
### Motivation / Current Behaviour
Introduction of flows in issue #511 and the added usage of templates to a lot of classes in issue #894 changed the behavior of the generation package. This results …
-
Original Repository: https://github.com/ml-explore/mlx-examples/
Listing out examples from there which would be nice to have. We don't expect the models to work out the moment they are translated to …