Doesn't run on CPU

[doevent]

Running on CPU, it gives an error:

Traceback (most recent call last):
  File "sample.py", line 554, in <module>
    do_run()
  File "sample.py", line 319, in do_run
    text_emb = bert.encode([args.text]*args.batch_size).to(device).float()
  File "/home/user/server/encoders/modules.py", line 102, in encode
    return self(text)
  File "/home/user/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/user/server/encoders/modules.py", line 94, in forward
    tokens = self.tknz_fn(text)#.to(self.device)
  File "/home/user/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/user/server/encoders/modules.py", line 65, in forward
    tokens = batch_encoding["input_ids"].to(self.device)
  File "/home/user/.local/lib/python3.8/site-packages/torch/cuda/__init__.py", line 216, in _lazy_init
    torch._C._cuda_init()
RuntimeError: Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.aspx

command:
python sample.py --model_path erlich.pt --batch_size 1 --num_batches 1 --text "a cyberpunk girl with a scifi neuralink device on her head" --cpu

[afiaka87]

@doevent

Noted - thanks will work on a fix soon hopefully. Although I should mention that this model is likely to take a very long time on CPU.

[alishan2040]

@afiaka87 My question is about GPU model ideal for training/fine-tuning latent-diffusion model. I had different GPUs with vram 16 to 24 GBs but still faced memory issues. What settings did you use to fine-tune latent diffusion with which GPU? Thanks.