@ancient-wolf-19325 why am I getting this error?

╭─────────────────────────────── Traceback (most recent call last) ────────────────────────────────╮
│   455 │   │   │   │   if self._state in [CANCELLED, CANCELLED_AND_NOTIFIED]:                     │
│   456 │   │   │   │   │   raise CancelledError()                                                 │
│   457 │   │   │   │   elif self._state == FINISHED:                                              │
│ ❱ 458 │   │   │   │   │   return self.__get_result()                                             │
│   459 │   │   │   │   else:                                                                      │
│   460 │   │   │   │   │   raise TimeoutError()                                                   │
│   461 │   │   finally:                                                                           │
│                                                                                                  │
│ __get_result                                                                                     │
│                                                                                                  │
│   400 │   def __get_result(self):                                                                │
│   401 │   │   if self._exception:                                                                │
│   402 │   │   │   try:                                                                           │
│ ❱ 403 │   │   │   │   raise self._exception                                                      │
│   404 │   │   │   finally:                                                                       │
│   405 │   │   │   │   # Break a reference cycle with the exception in self._exception            │
│   406 │   │   │   │   self = None                                                                │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
ValueError: cuda and cudnn do not need to be specified. If you are installed a GPU accelerated library on PyPI, then it likely will install cuda from PyPI.With conda you can installed cuda from the `nvidia` channel by adding `nvidia` to ImageSpec.conda_channels and 
adding packages from <https://anaconda.org/nvidia> into ImageSpec.conda_packages. If you require cuda for non-python dependencies, you can set a `base_image` with cuda preinstalled.

You are getting this error because the ImageSpec in Flytekit does not require the cuda and cudnn parameters to be set directly for most use cases. If you need CUDA for Python libraries (like torch or tensorflow), those libraries typically install the necessary CUDA dependencies themselves. For conda-based installs, you should add nvidia to conda_channels and specify CUDA packages from the NVIDIA channel in conda_packages. If you need CUDA for non-Python dependencies, set a base_image that already has CUDA preinstalled. See the error message and ImageSpec implementation for details. Would you like more detail on how to correctly configure GPU dependencies in Flyte ImageSpec? Sources: - flytekit/image_spec/default_builder.py 👍 0 button 👎 0 button 💡 Hint: Mention @runllm in the thread for followups.