Unable to use my model after training

[GamerGuy95953]

Describe the bug:

When I go to train, set my parameters, and start the train, when it finishes it says it is completed successfully but my model does not show up in the model inference menu. The only thing I could find that seemed to contain the results is in the logs folder. I've been suffering with RVC issues for the past 2 years and finally got the program to do something without major errors or crashes but I need help finding and using the model I created. I have attached the log and a PDF of what my parameters were set to.

Any help is appreciated!

To Reproduce:

Steps to reproduce the behavior:

1. go to "train" tab
2. set parameters such as training folder location, batch size, etc.
3. click on "one-click training"

Expected behavior:

The training is complete and the model shows up in the Inferencing voice dropdown

Actual behavior:

The training is complete but I cannot find the model and it does not show up in the Inferencing voice dropdown.

Desktop:

• OS and version: Ubuntu 24.04.1 LTS
• Python version: 3.10.16
• Commit/Tag with the issue: [e.g. 22]

Additional info:

log.txt
console_log.txt
RVC WebUI.pdf

[blueskies001029]

I'm using machine translation. Sorry if my sentences don't make sense.
The '.index' file is stored in the logs folder and the '.pth' file is stored in the assets/weights folder. The '.index' file doesn't exist in the logs folder you showed, so the model training didn't actually work!
To find out the exact cause of the problem, try running [Process data] -> [Feature extraction] -> [Train index file] -> [Train model]. One-click training is easy, but it doesn't show some errors.

[GamerGuy95953]

I'm using machine translation. Sorry if my sentences don't make sense. The '.index' file is stored in the logs folder and the '.pth' file is stored in the assets/weights folder. The '.index' file doesn't exist in the logs folder you showed, so the model training didn't actually work! To find out the exact cause of the problem, try running [Process data] -> [Feature extraction] -> [Train index file] -> [Train model]. One-click training is easy, but it doesn't show some errors.

Ok, it seems to get stuck on feature extraction. It loads the hubert_base.pt file but no disk or memory activity is happening at this step. I have attached an updated log. Right now I have still left it running so maybe it takes a super long time without any signs of resource usage?

Log 2.txt

[GamerGuy95953]

I'm using machine translation. Sorry if my sentences don't make sense. The '.index' file is stored in the logs folder and the '.pth' file is stored in the assets/weights folder. The '.index' file doesn't exist in the logs folder you showed, so the model training didn't actually work! To find out the exact cause of the problem, try running [Process data] -> [Feature extraction] -> [Train index file] -> [Train model]. One-click training is easy, but it doesn't show some errors.

Ok, it seems to get stuck on feature extraction. It loads the hubert_base.pt file but no disk or memory activity is happening at this step. I have attached an updated log. Right now I have still left it running so maybe it takes a super long time without any signs of resource usage?

Log 2.txt

Just in case it was actually finished, I did train model next as train feature index did nothing and attached is what I got.

Log 2 part 2.txt

[fumiama]

Maybe an AMD-GPU-related issue.

[GamerGuy95953]

Maybe an AMD-GPU-related issue.

Would the RX 7000 series work better?

[GamerGuy95953]

Ok, so I have two 7600 XT's now, and in the log, it says it detected at least one but in the WebUI, now it says that no compatible GPU's available to support my training.