No GPU Available in Pytorch + PytorchStreamReader failed #161

The issue is, Pytorch (2.0) detects my GPU (GTX 1660 Ti, CUDA 1.17) in every other executable, and with torch.cuda.is_available. It works with whisper as a standalone as well, just not this repo for some reason. During setup-cuda.bat runtime there is no errors, it conclude soundly. I editted the setup.bat to install 1.17 (which I already preinstalled). Any suggestions would be very helpful!

Stack trace:
!!!! WARNING !!!! No GPU available in PyTorch. You may need to reinstall PyTorch.
Loading TorToiSe... (AR: None, vocoder: bigvgan_24khz_100band)
No hardware acceleration is available, falling back to CPU...
Loading tokenizer JSON: C:\Users\ai-voice-cloning\modules\tortoise-tts\tortoise../tortoise/data/tokenizer.json
Loaded tokenizer
Loading autoregressive model: C:\Users\ai-voice-cloning\models\tortoise\autoregressive.pth
Loaded autoregressive model
Loaded diffusion model
Traceback (most recent call last):
File "C:\Users\ai-voice-cloning\src\main.py", line 22, in
tts = setup_tortoise()
File "C:\Users\ai-voice-cloning\src\utils.py", line 2347, in load_tts
tts = TextToSpeech(minor_optimizations=not args.low_vram, autoregressive_model_path=autoregressive_model, diffusion_model_path=diffusion_model, vocoder_model=vocoder_model, tokenizer_json=tokenizer_json)
File "c:\users\ai-voice-cloning\modules\tortoise-tts\tortoise\api.py", line 330, in init
self.clvp.load_state_dict(torch.load(get_model_path('clvp2.pth', models_dir)))
File "C:\Users\ai-voice-cloning\venv\lib\site-packages\torch\serialization.py", line 797, in load
with _open_zipfile_reader(opened_file) as opened_zipfile:
File "C:\Users\ai-voice-cloning\venv\lib\site-packages\torch\serialization.py", line 283, in init
super().init(torch._C.PyTorchFileReader(name_or_buffer))
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory
Press any key to continue . . .

The issue is, Pytorch (2.0) detects my GPU (GTX 1660 Ti, CUDA 1.17) in every other executable, and with torch.cuda.is_available. It works with whisper as a standalone as well, just not this repo for some reason. During setup-cuda.bat runtime there is no errors, it conclude soundly. I editted the setup.bat to install 1.17 (which I already preinstalled). Any suggestions would be very helpful! Stack trace: !!!! WARNING !!!! No GPU available in PyTorch. You may need to reinstall PyTorch. Loading TorToiSe... (AR: None, vocoder: bigvgan_24khz_100band) No hardware acceleration is available, falling back to CPU... Loading tokenizer JSON: C:\Users\\ai-voice-cloning\modules\tortoise-tts\tortoise\../tortoise/data/tokenizer.json Loaded tokenizer Loading autoregressive model: C:\Users\\ai-voice-cloning\models\tortoise\autoregressive.pth Loaded autoregressive model Loaded diffusion model Traceback (most recent call last): File "C:\Users\\ai-voice-cloning\src\main.py", line 22, in <module> tts = setup_tortoise() File "C:\Users\\ai-voice-cloning\src\utils.py", line 2347, in load_tts tts = TextToSpeech(minor_optimizations=not args.low_vram, autoregressive_model_path=autoregressive_model, diffusion_model_path=diffusion_model, vocoder_model=vocoder_model, tokenizer_json=tokenizer_json) File "c:\users\\ai-voice-cloning\modules\tortoise-tts\tortoise\api.py", line 330, in __init__ self.clvp.load_state_dict(torch.load(get_model_path('clvp2.pth', models_dir))) File "C:\Users\\ai-voice-cloning\venv\lib\site-packages\torch\serialization.py", line 797, in load with _open_zipfile_reader(opened_file) as opened_zipfile: File "C:\Users\\ai-voice-cloning\venv\lib\site-packages\torch\serialization.py", line 283, in __init__ super().__init__(torch._C.PyTorchFileReader(name_or_buffer)) RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory Press any key to continue . . .

I fixed this issue. File was corrupted. A clean install of CUDA + Pytorch + delete the repo and clone it again solved the problem.

Labels Milestones

No GPU Available in Pytorch + PytorchStreamReader failed #161