Error running, getting this log
Exit code: 3. Reason: se of pip is available: 24.3.1 -> 25.0.1
[notice] To update, run: python -m pip install --upgrade pip
ERROR: ResolutionImpossible: for help visit https://pip.pypa.io/en/latest/topics/dependency-resolution/#dealing-with-dependency-conflicts
2025-02-18 09:48:45 - huggingface_inference_toolkit - INFO - Initializing model from directory:/repository
2025-02-18 09:48:45 - huggingface_inference_toolkit - INFO - Found custom pipeline at /repository/handler.py
Traceback (most recent call last):
File "/usr/local/lib/python3.11/dist-packages/starlette/routing.py", line 693, in lifespan
async with self.lifespan_context(app) as maybe_state:
File "/usr/local/lib/python3.11/dist-packages/starlette/routing.py", line 569, in aenter
await self._router.startup()
File "/usr/local/lib/python3.11/dist-packages/starlette/routing.py", line 670, in startup
await handler()
File "/app/webservice_starlette.py", line 62, in prepare_model_artifacts
inference_handler = get_inference_handler_either_custom_or_default_handler(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/huggingface_inference_toolkit/handler.py", line 152, in get_inference_handler_either_custom_or_default_handler
custom_pipeline = check_and_register_custom_pipeline_from_directory(model_dir)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/huggingface_inference_toolkit/utils.py", line 168, in check_and_register_custom_pipeline_from_directory
spec.loader.exec_module(handler)
File "", line 940, in exec_module
File "", line 241, in _call_with_frames_removed
File "/repository/handler.py", line 9, in
from skyreelsinfer.offload import OffloadConfig
File "/repository/skyreelsinfer/offload.py", line 9, in
from torchao.dtypes.affine_quantized_tensor import AffineQuantizedTensor
ModuleNotFoundError: No module named 'torchao'
Application startup failed. Exiting.
Hello,
I've fixed this particular issues but there are more fixes needed, as this Image-to-Video uses some custom code (the current version Diffusers doesn't support Image-to-Video while this model does, so I need to look at how to integrate this)
-> I'm going to first attempt to make the Text-to-Video version work first, then I will go back to this one to investigate
Aah Ok, thanks