TypeError: OrthogonalRegularizer.__init__() got an unexpected keyword argument 'num_features'

[Allaye]

Describe the bug

I encountered a TypeError while trying to load the pointnet_segmentation model from Hugging Face using the from_pretrained_keras function. The error message indicates that the __init__ method of OrthogonalRegularizer does not accept the num_features argument.

Reproduction

Install the required packages

!pip install -U keras==2.12.0 huggingface_hub

Import the necessary library

from huggingface_hub import from_pretrained_keras

Attempting to load the PointNet segmentation model

model = from_pretrained_keras("keras-io/pointnet_segmentation")

Logs

TypeError: OrthogonalRegularizer.__init__() got an unexpected keyword argument 'num_features'

TypeError                                 Traceback (most recent call last)
<ipython-input-4-e8c7f05e3ba1> in <cell line: 4>()
      2 # !pip install -U keras==2.12.0 huggingface_hub
      3 from huggingface_hub import from_pretrained_keras
----> 4 model = from_pretrained_keras("keras-io/pointnet_segmentation")
      5 #Wmodel.summary("keras/")

6 frames
/usr/local/lib/python3.10/dist-packages/tf_keras/src/regularizers.py in from_config(cls, config)
    188             A regularizer instance.
    189         """
--> 190         return cls(**config)
    191 
    192     def get_config(self):

TypeError: OrthogonalRegularizer.__init__() got an unexpected keyword argument 'num_features'

System info

- tensorboard: N/A
- numpy: 1.26.4
- pydantic: 2.9.2
- aiohttp: 3.11.2
- ENDPOINT: https://huggingface.co
- HF_HUB_CACHE: /root/.cache/huggingface/hub
- HF_ASSETS_CACHE: /root/.cache/huggingface/assets
- HF_TOKEN_PATH: /root/.cache/huggingface/token
- HF_STORED_TOKENS_PATH: /root/.cache/huggingface/stored_tokens
- HF_HUB_OFFLINE: False
- HF_HUB_DISABLE_TELEMETRY: False
- HF_HUB_DISABLE_PROGRESS_BARS: None
- HF_HUB_DISABLE_SYMLINKS_WARNING: False
- HF_HUB_DISABLE_EXPERIMENTAL_WARNING: False
- HF_HUB_DISABLE_IMPLICIT_TOKEN: False
- HF_HUB_ENABLE_HF_TRANSFER: False
- HF_HUB_ETAG_TIMEOUT: 10
- HF_HUB_DOWNLOAD_TIMEOUT: 10

{'huggingface_hub version': '0.26.2',
 'Platform': 'Linux-6.1.85+-x86_64-with-glibc2.35',
 'Python version': '3.10.12',
 'Running in iPython ?': 'Yes',
 'iPython shell': 'Shell',
 'Running in notebook ?': 'Yes',
 'Running in Google Colab ?': 'Yes',
 'Running in Google Colab Enterprise ?': 'No',
 'Token path ?': '/root/.cache/huggingface/token',
 'Has saved token ?': False,
 'Configured git credential helpers': '',
 'FastAI': '2.7.18',
 'Tensorflow': '2.17.1',
 'Torch': '2.5.1+cu121',
 'Jinja2': '3.1.4',
 'Graphviz': '0.20.3',
 'keras': '3.5.0',
 'Pydot': '3.0.2',
 'Pillow': '11.0.0',
 'hf_transfer': 'N/A',
 'gradio': 'N/A',
 'tensorboard': 'N/A',
 'numpy': '1.26.4',
 'pydantic': '2.9.2',
 'aiohttp': '3.11.2',
 'ENDPOINT': 'https://huggingface.co',
 'HF_HUB_CACHE': '/root/.cache/huggingface/hub',
 'HF_ASSETS_CACHE': '/root/.cache/huggingface/assets',
 'HF_TOKEN_PATH': '/root/.cache/huggingface/token',
 'HF_STORED_TOKENS_PATH': '/root/.cache/huggingface/stored_tokens',
 'HF_HUB_OFFLINE': False,
 'HF_HUB_DISABLE_TELEMETRY': False,
 'HF_HUB_DISABLE_PROGRESS_BARS': None,
 'HF_HUB_DISABLE_SYMLINKS_WARNING': False,
 'HF_HUB_DISABLE_EXPERIMENTAL_WARNING': False,
 'HF_HUB_DISABLE_IMPLICIT_TOKEN': False,
 'HF_HUB_ENABLE_HF_TRANSFER': False,
 'HF_HUB_ETAG_TIMEOUT': 10,
 'HF_HUB_DOWNLOAD_TIMEOUT': 10}

[Wauplin]

Hi @Allaye , I'm sorry you're facing this issue. This error doesn't seem related to huggingface_hub itself but to an incompatibility between the model and the keras/tensorflow version. The error occurs while loading the model from the downloaded file using the keras method. What I would advice is to open an issue on the model Discussion tab: https://huggingface.co/keras-io/pointnet_segmentation/discussions. Note that I'm not sure these models are really well maintained given they are ~3 years old and the Keras got heavily updated since then.

I'll close this issue but let me know if you have any further question.