--- language: - multilingual - en - de - fr - ja license: mit tags: - object-detection - vision - generated_from_trainer - DocLayNet - COCO - PDF - IBM - Financial-Reports - Finance - Manuals - Scientific-Articles - Science - Laws - Law - Regulations - Patents - Government-Tenders - object-detection - image-segmentation - token-classification inference: false datasets: - pierreguillou/DocLayNet-base metrics: - precision - recall - f1 - accuracy model-index: - name: lilt-xlm-roberta-base-finetuned-with-DocLayNet-base-at-paragraphlevel-ml512 results: - task: name: Token Classification type: token-classification metrics: - name: f1 type: f1 value: 0.8634 - name: accuracy type: accuracy value: 0.6815 --- # Document Understanding model (finetuned LiLT base at paragraph level on DocLayNet base) This model is a fine-tuned version of [nielsr/lilt-xlm-roberta-base](https://huggingface.co/nielsr/lilt-xlm-roberta-base) with the [DocLayNet base](https://huggingface.co/datasets/pierreguillou/DocLayNet-base) dataset. It achieves the following results on the evaluation set: - Loss: 0.4104 - Precision: 0.8634 - Recall: 0.8634 - F1: 0.8634 - Token Accuracy: 0.8634 - Paragraph Accuracy: 0.6815 ## Accuracy at paragraph level - Paragraph Accuracy: 68.15% - Accuracy by label - Caption: 22.82% - Footnote: 0.0% - Formula: 97.33% - List-item: 8.42% - Page-footer: 98.77% - Page-header: 77.81% - Picture: 39.16% - Section-header: 76.17% - Table: 37.7% - Text: 86.78% - Title: 0.0% ![Paragraphs labels vs accuracy (%) of the dataset DocLayNet base of test (model: LiLT base finetuned on DocLayNet base))](https://huggingface.co/pierreguillou/lilt-xlm-roberta-base-finetuned-with-DocLayNet-base-at-paragraphlevel-ml512/resolve/main/docs/paragraphs_labels_accuracy_DocLayNet_base_test_LiLT_base_paragraph_level_512.png) ![Confusion matrix of the labeled blocks of the dataset DocLayNet base of test (model: LiLT base finetuned on DocLayNet base)](https://huggingface.co/pierreguillou/lilt-xlm-roberta-base-finetuned-with-DocLayNet-base-at-paragraphlevel-ml512/resolve/main/docs/confusion_matrix_labeled_paragraphs_DocLayNet_base_test_LiLT_base_paragraph_level_512.png) ## References ### Other model - LayoutXLM base - [Document Understanding model (at line level)](https://huggingface.co/pierreguillou/layout-xlm-base-finetuned-with-DocLayNet-base-at-linelevel-ml384) - LiLT base - [Document Understanding model (at paragraph level)](https://huggingface.co/pierreguillou/lilt-xlm-roberta-base-finetuned-with-DocLayNet-base-at-paragraphlevel-ml512) - [Document Understanding model (at line level)](https://huggingface.co/pierreguillou/lilt-xlm-roberta-base-finetuned-with-DocLayNet-base-at-linelevel-ml384) ### Blog posts - Layout XLM base - (03/05/2023) [Document AI | Inference APP and fine-tuning notebook for Document Understanding at line level with LayoutXLM base]() - LiLT base - (02/16/2023) [Document AI | Inference APP and fine-tuning notebook for Document Understanding at paragraph level](https://medium.com/@pierre_guillou/document-ai-inference-app-and-fine-tuning-notebook-for-document-understanding-at-paragraph-level-c18d16e53cf8) - (02/14/2023) [Document AI | Inference APP for Document Understanding at line level](https://medium.com/@pierre_guillou/document-ai-inference-app-for-document-understanding-at-line-level-a35bbfa98893) - (02/10/2023) [Document AI | Document Understanding model at line level with LiLT, Tesseract and DocLayNet dataset](https://medium.com/@pierre_guillou/document-ai-document-understanding-model-at-line-level-with-lilt-tesseract-and-doclaynet-dataset-347107a643b8) - (01/31/2023) [Document AI | DocLayNet image viewer APP](https://medium.com/@pierre_guillou/document-ai-doclaynet-image-viewer-app-3ac54c19956) - (01/27/2023) [Document AI | Processing of DocLayNet dataset to be used by layout models of the Hugging Face hub (finetuning, inference)](https://medium.com/@pierre_guillou/document-ai-processing-of-doclaynet-dataset-to-be-used-by-layout-models-of-the-hugging-face-hub-308d8bd81cdb) ### Notebooks (paragraph level) - LiLT base - [Document AI | Inference APP at paragraph level with a Document Understanding model (LiLT fine-tuned on DocLayNet dataset)](https://github.com/piegu/language-models/blob/master/Gradio_inference_on_LiLT_model_finetuned_on_DocLayNet_base_in_any_language_at_levelparagraphs_ml512.ipynb) - [Document AI | Inference at paragraph level with a Document Understanding model (LiLT fine-tuned on DocLayNet dataset)](https://github.com/piegu/language-models/blob/master/inference_on_LiLT_model_finetuned_on_DocLayNet_base_in_any_language_at_levelparagraphs_ml512.ipynb) - [Document AI | Fine-tune LiLT on DocLayNet base in any language at paragraph level (chunk of 512 tokens with overlap)](https://github.com/piegu/language-models/blob/master/Fine_tune_LiLT_on_DocLayNet_base_in_any_language_at_paragraphlevel_ml_512.ipynb) ### Notebooks (line level) - Layout XLM base - [Document AI | Inference at line level with a Document Understanding model (LayoutXLM base fine-tuned on DocLayNet dataset)](https://github.com/piegu/language-models/blob/master/inference_on_LayoutXLM_base_model_finetuned_on_DocLayNet_base_in_any_language_at_levellines_ml384.ipynb) - [Document AI | Inference APP at line level with a Document Understanding model (LayoutXLM base fine-tuned on DocLayNet base dataset)](https://github.com/piegu/language-models/blob/master/Gradio_inference_on_LayoutXLM_base_model_finetuned_on_DocLayNet_base_in_any_language_at_levellines_ml384.ipynb) - [Document AI | Fine-tune LayoutXLM base on DocLayNet base in any language at line level (chunk of 384 tokens with overlap)](https://github.com/piegu/language-models/blob/master/Fine_tune_LayoutXLM_base_on_DocLayNet_base_in_any_language_at_linelevel_ml_384.ipynb) - LiLT base - [Document AI | Inference at line level with a Document Understanding model (LiLT fine-tuned on DocLayNet dataset)](https://github.com/piegu/language-models/blob/master/inference_on_LiLT_model_finetuned_on_DocLayNet_base_in_any_language_at_levellines_ml384.ipynb) - [Document AI | Inference APP at line level with a Document Understanding model (LiLT fine-tuned on DocLayNet dataset)](https://github.com/piegu/language-models/blob/master/Gradio_inference_on_LiLT_model_finetuned_on_DocLayNet_base_in_any_language_at_levellines_ml384.ipynb) - [Document AI | Fine-tune LiLT on DocLayNet base in any language at line level (chunk of 384 tokens with overlap)](https://github.com/piegu/language-models/blob/master/Fine_tune_LiLT_on_DocLayNet_base_in_any_language_at_linelevel_ml_384.ipynb) - [DocLayNet image viewer APP](https://github.com/piegu/language-models/blob/master/DocLayNet_image_viewer_APP.ipynb) - [Processing of DocLayNet dataset to be used by layout models of the Hugging Face hub (finetuning, inference)](processing_DocLayNet_dataset_to_be_used_by_layout_models_of_HF_hub.ipynb) ## APP You can test this model with this APP in Hugging Face Spaces: [Inference APP for Document Understanding at paragraph level (v1)](https://huggingface.co/spaces/pierreguillou/Inference-APP-Document-Understanding-at-paragraphlevel-v1). ![Inference APP for Document Understanding at paragraph level (v1)](https://huggingface.co/pierreguillou/lilt-xlm-roberta-base-finetuned-with-DocLayNet-base-at-paragraphlevel-ml512/resolve/main/docs/app_lilt_document_understanding_AI_paragraphlevel.png) You can run as well the corresponding notebook: [Document AI | Inference APP at paragraph level with a Document Understanding model (LiLT fine-tuned on DocLayNet dataset)](https://github.com/piegu/language-models/blob/master/Gradio_inference_on_LiLT_model_finetuned_on_DocLayNet_base_in_any_language_at_levelparagraphs_ml512.ipynb) ## DocLayNet dataset [DocLayNet dataset](https://github.com/DS4SD/DocLayNet) (IBM) provides page-by-page layout segmentation ground-truth using bounding-boxes for 11 distinct class labels on 80863 unique pages from 6 document categories. Until today, the dataset can be downloaded through direct links or as a dataset from Hugging Face datasets: - direct links: [doclaynet_core.zip](https://codait-cos-dax.s3.us.cloud-object-storage.appdomain.cloud/dax-doclaynet/1.0.0/DocLayNet_core.zip) (28 GiB), [doclaynet_extra.zip](https://codait-cos-dax.s3.us.cloud-object-storage.appdomain.cloud/dax-doclaynet/1.0.0/DocLayNet_extra.zip) (7.5 GiB) - Hugging Face dataset library: [dataset DocLayNet](https://huggingface.co/datasets/ds4sd/DocLayNet) Paper: [DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis](https://arxiv.org/abs/2206.01062) (06/02/2022) ## Model description The model was finetuned at **paragraph level on chunk of 512 tokens with overlap of 128 tokens**. Thus, the model was trained with all layout and text data of all pages of the dataset. At inference time, a calculation of best probabilities give the label to each paragraph bounding boxes. ## Inference See notebook: [Document AI | Inference at paragraph level with a Document Understanding model (LiLT fine-tuned on DocLayNet dataset)](https://github.com/piegu/language-models/blob/master/inference_on_LiLT_model_finetuned_on_DocLayNet_base_in_any_language_at_levelparagraphs_ml512.ipynb) ## Training and evaluation data See notebook: [Document AI | Fine-tune LiLT on DocLayNet base in any language at paragraph level (chunk of 512 tokens with overlap)](https://github.com/piegu/language-models/blob/master/Fine_tune_LiLT_on_DocLayNet_base_in_any_language_at_paragraphlevel_ml_512.ipynb) ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| | No log | 0.05 | 100 | 0.9875 | 0.6585 | 0.6585 | 0.6585 | 0.6585 | | No log | 0.11 | 200 | 0.7886 | 0.7551 | 0.7551 | 0.7551 | 0.7551 | | No log | 0.16 | 300 | 0.5894 | 0.8248 | 0.8248 | 0.8248 | 0.8248 | | No log | 0.21 | 400 | 0.4794 | 0.8396 | 0.8396 | 0.8396 | 0.8396 | | 0.7446 | 0.27 | 500 | 0.3993 | 0.8703 | 0.8703 | 0.8703 | 0.8703 | | 0.7446 | 0.32 | 600 | 0.3631 | 0.8857 | 0.8857 | 0.8857 | 0.8857 | | 0.7446 | 0.37 | 700 | 0.4096 | 0.8630 | 0.8630 | 0.8630 | 0.8630 | | 0.7446 | 0.43 | 800 | 0.4492 | 0.8528 | 0.8528 | 0.8528 | 0.8528 | | 0.7446 | 0.48 | 900 | 0.3839 | 0.8834 | 0.8834 | 0.8834 | 0.8834 | | 0.4464 | 0.53 | 1000 | 0.4365 | 0.8498 | 0.8498 | 0.8498 | 0.8498 | | 0.4464 | 0.59 | 1100 | 0.3616 | 0.8812 | 0.8812 | 0.8812 | 0.8812 | | 0.4464 | 0.64 | 1200 | 0.3949 | 0.8796 | 0.8796 | 0.8796 | 0.8796 | | 0.4464 | 0.69 | 1300 | 0.4184 | 0.8613 | 0.8613 | 0.8613 | 0.8613 | | 0.4464 | 0.75 | 1400 | 0.4130 | 0.8743 | 0.8743 | 0.8743 | 0.8743 | | 0.3672 | 0.8 | 1500 | 0.4535 | 0.8289 | 0.8289 | 0.8289 | 0.8289 | | 0.3672 | 0.85 | 1600 | 0.3681 | 0.8713 | 0.8713 | 0.8713 | 0.8713 | | 0.3672 | 0.91 | 1700 | 0.3446 | 0.8857 | 0.8857 | 0.8857 | 0.8857 | | 0.3672 | 0.96 | 1800 | 0.4104 | 0.8634 | 0.8634 | 0.8634 | 0.8634 | ### Framework versions - Transformers 4.26.1 - Pytorch 1.13.1+cu116 - Datasets 2.9.0 - Tokenizers 0.13.2 ## Other models - Line level - [Document Understanding model (finetuned LiLT base at line level on DocLayNet base)](https://huggingface.co/pierreguillou/lilt-xlm-roberta-base-finetuned-with-DocLayNet-base-at-linelevel-ml384) (line accuracy: xxxx) - [Document Understanding model (finetuned LayoutXLM base at line level on DocLayNet base)](https://huggingface.co/pierreguillou/layout-xlm-base-finetuned-with-DocLayNet-base-at-linelevel-ml384) (line accuracy: xxx) - Paragraph level - [Document Understanding model (finetuned LiLT base at paragraph level on DocLayNet base)](https://huggingface.co/pierreguillou/lilt-xlm-roberta-base-finetuned-with-DocLayNet-base-at-paragraphlevel-ml512) (paragraph accuracy: 68.15%) - [Document Understanding model (finetuned LayoutXLM base at paragraph level on DocLayNet base)](https://huggingface.co/pierreguillou/layout-xlm-base-finetuned-with-DocLayNet-base-at-paragraphlevel-ml512) (paragraph accuracy: 86.55%)