Error on fresh install starting training

[dsienra]

grad_accum: 1 batch_size: 1 epoch_len: 176 Epochs: 0%| | 0/200 [00:00<?, ?it/s, vram=5509/12288 MB gs:0]C:\EveryDream2trainer\venv\lib\site-packages\xformers\ops\fmha\flash.py:339: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() and inp.query.storage().data_ptr() == inp.key.storage().data_ptr() Something went wrong, attempting to save model No model to save, something likely blew up on startup, not saving Traceback (most recent call last): File "C:\EveryDream2trainer\train.py", line 916, in main(args) File "C:\EveryDream2trainer\train.py", line 844, in main raise ex File "C:\EveryDream2trainer\train.py", line 756, in main ed_optimizer.step(loss, step, global_step) File "C:\EveryDream2trainer\optimizer\optimizers.py", line 98, in step self.scaler.step(optimizer) File "C:\EveryDream2trainer\venv\lib\site-packages\torch\cuda\amp\grad_scaler.py", line 374, in step retval = self._maybe_opt_step(optimizer, optimizer_state, *args, kwargs) File "C:\EveryDream2trainer\venv\lib\site-packages\torch\cuda\amp\grad_scaler.py", line 290, in _maybe_opt_step retval = optimizer.step(*args, *kwargs) File "C:\EveryDream2trainer\venv\lib\site-packages\torch\optim\lr_scheduler.py", line 69, in wrapper return wrapped(args, kwargs) File "C:\EveryDream2trainer\venv\lib\site-packages\torch\optim\optimizer.py", line 280, in wrapper out = func(*args, kwargs) File "C:\EveryDream2trainer\venv\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context return func(*args, *kwargs) File "C:\EveryDream2trainer\venv\lib\site-packages\bitsandbytes\optim\optimizer.py", line 263, in step self.update_step(group, p, gindex, pindex) File "C:\EveryDream2trainer\venv\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context return func(args, kwargs) File "C:\EveryDream2trainer\venv\lib\site-packages\bitsandbytes\optim\optimizer.py", line 504, in update_step F.optimizer_update_8bit_blockwise( File "C:\EveryDream2trainer\venv\lib\site-packages\bitsandbytes\functional.py", line 981, in optimizer_update_8bit_blockwise str2optimizer8bit_blockwise[optimizer_name][0]( NameError: name 'str2optimizer8bit_blockwise' is not defined Epochs: 0%| | 0/200 [00:11<?, ?it/s, vram=5509/12288 MB gs:0]

(venv) C:\EveryDream2trainer>

[victorchall]

Try installing cuda 11.8

https://developer.nvidia.com/cuda-11-8-0-download-archive

Make sure to start a fresh command window after that, activate, then try again.

[wavymulder]

Try installing cuda 11.8

I was encountering the same issue and this resolved it for me. Thanks Victor.

[victorchall]

@dsienra is this still an issue? Did you try installing cuda 11.8 from the link?

[victorchall]

Pretty sure this will correct the error. Going to close. If this is still a problem please open a new issue.