search

IAHispano / Applio

VITS-based Voice Conversion focused on simplicity, quality and performance.
https://applio.org
MIT License
1.36k stars 230 forks source link

[BUG] #411

Closed edmundman closed 2 months ago

edmundman commented 3 months ago

Bug Description Error cuda out of mem when trying to train a model but i have a 3090 f0D40k.pth

Process Process-1: Traceback (most recent call last): File "/home/ed/mambaforge/envs/applio/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/ed/mambaforge/envs/applio/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/ed/Applio/rvc/train/train.py", line 258, in run train_and_evaluate( File "/home/ed/Applio/rvc/train/train.py", line 457, in train_and_evaluate scaler.scale(loss_disc).backward() File "/home/ed/mambaforge/envs/applio/lib/python3.10/site-packages/torch/_tensor.py", line 492, in backward torch.autograd.backward( File "/home/ed/mambaforge/envs/applio/lib/python3.10/site-packages/torch/autograd/__init__.py", line 251, in backward Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass RuntimeError: CUDA error: out of memory CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1. Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. **Desktop Details:** - Operating System: [WSL UBUNTU ] - Browser: [Firefox]
blaisewf commented 2 months ago

You have Windows? Why you train on a WSL