Instructions to use HYPJUDY/layoutlmv3-base-finetuned-publaynet with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use HYPJUDY/layoutlmv3-base-finetuned-publaynet with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("HYPJUDY/layoutlmv3-base-finetuned-publaynet", dtype="auto") - Notebooks
- Google Colab
- Kaggle
preprocessor_config.json file absent
Following is my Code Snippet
########################### CODE START #########################################
from transformers import AutoProcessor, AutoModelForTokenClassification
MODEL_ID = "HYPJUDY/layoutlmv3-base-finetuned-publaynet"
processor = AutoProcessor.from_pretrained(model_id, apply_ocr=False)
model = AutoModelForTokenClassification.from_pretrained(model_id).to(device)
########################### CODE END #########################################
########################### ERROR START #########################################
This gives the error :- HYPJUDY/layoutlmv3-base-finetuned-publaynet does not appear to have a file named preprocessor_config.json. Checkout 'https://huggingface.co/HYPJUDY/layoutlmv3-base-finetuned-publaynet/tree/main'for available files.
########################### ERROR END #########################################
I checked the Files folder and having preprocessor_config file is usually present for other Hugging Face models.
Is direct preprocessing and inferencing not possible for this model using transformers package? My use case is specifically for model to understand the layout and output the information in structured JSON or a predefined structured summary.
Thank You!
this model does not have preprocessor_config.json
tokenizer.json
tokenizer_config.json
config.json
pytorch_model.bin so we can't use it at all
okay I made some research and if you need another project with all the file : go here
Theivaprakasham
/
layoutlmv3-finetuned-invoice
Thanks a lot for that. Will have a look at it for sure.