Error: `IndexError: list index out of range` in LoRAW Gradio Interface When Loading LoRA Model

[furmanlukasz]

Issue Description:

I encountered an error while running a trained LoRA checkpoint in the Gradio interface for LoRAW. The model loads successfully, but when attempting to generate audio, it throws an IndexError: list index out of range.

Error Log:

/workspace
Creating model from config
No module named 'flash_attn'
flash_attn not installed, disabling Flash Attention
/usr/local/lib/python3.10/dist-packages/torch/nn/utils/weight_norm.py:30: UserWarning: torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.
  warnings.warn("torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.")
Loading model checkpoint from /workspace/model.ckpt
Done loading model
Found 170 candidates for LoRA replacement
LoRA registered: stable_audio_open_finetune_loraw1
Running on local URL:  http://127.0.0.1:7860
Running on public URL: https://b59f39fd94fc1607b4.gradio.live

This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)
Prompt: ttett
Traceback (most recent call last):
  File "/usr/local/lib/python3.10/dist-packages/gradio/queueing.py", line 536, in process_events
    response = await route_utils.call_process_api(
  File "/usr/local/lib/python3.10/dist-packages/gradio/route_utils.py", line 321, in call_process_api
    output = await app.get_blocks().process_api(
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1935, in process_api
    result = await self.call_function(
  File "/usr/local/lib/python3.10/dist-packages/gradio/blocks.py", line 1520, in call_function
    prediction = await anyio.to_thread.run_sync(  # type: ignore
  File "/usr/local/lib/python3.10/dist-packages/anyio/to_thread.py", line 33, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
  File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 2106, in run_sync_in_worker_thread
    return await future
  File "/usr/local/lib/python3.10/dist-packages/anyio/_backends/_asyncio.py", line 833, in run
    result = context.run(func, *args)
  File "/usr/local/lib/python3.10/dist-packages/gradio/utils.py", line 826, in wrapper
    response = f(*args, **kwargs)
  File "/workspace/LoRAW/interface/gradio.py", line 175, in generate_cond
    lora_merger.merge({lora_names[0]: lora1, lora_names[1]: lora2, lora_names[2]: lora3})
IndexError: list index out of range

Steps to Reproduce:

1. Run the LoRAW interface with the following command:

   %cd /workspace
   command = (
      "python LoRAW/run_gradio.py"
      " --model-config /workspace/model_config_loraw.json"
      " --ckpt-path /workspace/model.ckpt"
      " --lora-dir /workspace/Lora-Model"
   )
   !{command}

2. The model successfully loads, and LoRA candidates are registered.
3. On submitting a text prompt, the error IndexError: list index out of range is triggered at gradio.py:175.

Additional Context:

• I am using a pre-trained LoRA checkpoint with a .ckpt file.
• Should I unwrap the model similar to how it's done with a vanilla model, or is it correct to load the checkpoint directly since it's already in the .ckpt format?

Any help resolving this issue or guidance on properly handling LoRA model checkpoints in this interface would be appreciated.

[NeuralNotW0rk]

Hey there, Gradio can't do variable number of components so when I hard-coded for up to 3 loras I forgot to check for fewer than 3. Fixed now.