Keras
Joblib
Safetensors
PyTorch
English
multimodal
document-classification
distilbert
efficientnet
late-fusion
rvl-cdip
tensorflow
scikit-learn
Instructions to use gonzalobosque/multimodal-document-intelligence with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use gonzalobosque/multimodal-document-intelligence with Keras:
# !pip install -U keras tensorflow huggingface_hub # Keras needs TensorFlow installed to read "hf://" paths, so the tensorflow backend is selected here; # "jax" and "torch" also work for computation once TensorFlow is installed. import os os.environ["KERAS_BACKEND"] = "tensorflow" import keras model = keras.saving.load_model("hf://gonzalobosque/multimodal-document-intelligence") - Notebooks
- Google Colab
- Kaggle
File size: 261 Bytes
84c9ebc | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 | [
"letter",
"form",
"email",
"handwritten",
"advertisement",
"scientific report",
"scientific publication",
"specification",
"file folder",
"news article",
"budget",
"invoice",
"presentation",
"questionnaire",
"resume",
"memo"
]
|