Да, я писал, по другому вопросу. Не ответили.
В логах такое написано. Да карта старая, но пишут что с gtx750 уже должно работать. Пробовал скачать более старую версию tonex 1,0,4 . Тоже ошибка.
Дрова обновил. Куду скачал... Не помогает.
--- System Information ---
[OS Information]
- System: Windows 10
- Architecture: AMD64
[CPU & Software Information]
- Processor: Intel(R) Core(TM) i7-6700K CPU @ 4.00GHz
- Python Version: 3.12.5
[GPU & PyTorch Backend Information]
- PyTorch Backend: CUDA is available
- CUDA Version (PyTorch): 12.9
- Detected CUDA Devices: 1
- Device 0: NVIDIA GeForce GTX 970
[NVIDIA GPU 0 Details]
- Name: NVIDIA GeForce GTX 970
- Driver Version: 581.29
- Total Memory: 4096 MiB
--- End of System Information ---
- Model Info -
Model name: 250913T102947_0
Training will rely on : cuda
- Training Info -
Retrained model from None
Detected latency samples: 1044
Parameter 1: 0.0
Parameter 2: 1288
ERROR (NetworkMain): Unhandled exception: CUDA error: no kernel image is available for execution on the device
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.