Chapter #9- Running issue

[armin55fp]

Hi,

When I am running the following part in the colab:

# Generate image ids to be passed to the decoder (LLM)
generated_ids = model.generate(**inputs, max_new_tokens=20)

# Generate text from the image ids
generated_text = blip_processor.batch_decode(generated_ids, skip_special_tokens=True)
generated_text = generated_text[0].strip()
generated_text

I am getting this error:

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
[<ipython-input-25-4e509c2e1006>](https://localhost:8080/#) in <cell line: 0>()
      1 # Generate image ids to be passed to the decoder (LLM)
----> 2 generated_ids = model.generate(**inputs, max_new_tokens=20)
      3 
      4 # Generate text from the image ids
      5 generated_text = blip_processor.batch_decode(generated_ids, skip_special_tokens=True)

8 frames
[/usr/local/lib/python3.11/dist-packages/torch/nn/functional.py](https://localhost:8080/#) in layer_norm(input, normalized_shape, weight, bias, eps)
   2908             eps=eps,
   2909         )
-> 2910     return torch.layer_norm(
   2911         input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled
   2912     )

RuntimeError: expected scalar type Float but found Half

Can you please explain?

[MaartenGr]

I haven't seen this issue before. Just to be sure, you ran the code in Google Colab without any changes?