icdar23-entrydetector_plaintext_breaks_indents_left_ref

HueyNemud/icdar23-entrydetector_plaintext_breaks_indents_left_ref

icdar23-entrydetector_plaintext_breaks_indents_left_ref is a token classification model from HueyNemud. Use it when you need labels on individual words, such as names. It is set up for transformers.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

17% of all-time downloads

All-time downloads

Public

Repo size

3.5 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin440 MB · 99%

At a glance

Task: Token Classification
Library: transformers
Model type: camembert
Access: Public
Created: Feb 8, 2023
Updated: Feb 10, 2023
SHA: 4c7ee801

Try a prompt

Model card

From the Hugging Face model README

icdar23-entrydetector_plaintext_breaks_indents_left_ref

This model is a fine-tuned version of HueyNemud/das22-10-camembert_pretrained on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0062
Ebegin: {'precision': 0.997709049255441, 'recall': 0.9827002632568634, 'f1': 0.9901477832512315, 'number': 2659}
Eend: {'precision': 0.9973363774733638, 'recall': 0.9794469357249627, 'f1': 0.9883107088989442, 'number': 2676}
Overall Precision: 0.9975
Overall Recall: 0.9811
Overall F1: 0.9892
Overall Accuracy: 0.9982

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0001
train_batch_size: 2
eval_batch_size: 2
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 7500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.07	300	0.0389	0.9556	0.9663	0.9609	0.9949
0.1775	0.14	600	0.0162	0.9854	0.9893	0.9873	0.9977
0.1775	0.21	900	0.0114	0.9928	0.9889	0.9909	0.9984
0.0229	0.29	1200	0.0172	0.9793	0.9851	0.9822	0.9975
0.016	0.36	1500	0.0087	0.9906	0.9907	0.9907	0.9984
0.016	0.43	1800	0.0079	0.9955	0.9879	0.9917	0.9985
0.0115	0.5	2100	0.0093	0.9910	0.9912	0.9911	0.9984
0.0115	0.57	2400	0.0102	0.9816	0.9942	0.9878	0.9978
0.0109	0.64	2700	0.0072	0.9895	0.9939	0.9917	0.9985
0.0075	0.72	3000	0.0055	0.9919	0.9917	0.9918	0.9985
0.0075	0.79	3300	0.0078	0.9948	0.9910	0.9929	0.9987
0.007	0.86	3600	0.0057	0.9937	0.9933	0.9935	0.9989
0.007	0.93	3900	0.0059	0.9830	0.9957	0.9893	0.9981
0.0055	1.0	4200	0.0049	0.9972	0.9899	0.9935	0.9988
0.0029	1.07	4500	0.0064	0.9944	0.9926	0.9935	0.9989
0.0029	1.14	4800	0.0057	0.9927	0.9919	0.9923	0.9987
0.0043	1.22	5100	0.0064	0.9890	0.9945	0.9917	0.9986
0.0043	1.29	5400	0.0058	0.9857	0.9957	0.9907	0.9983
0.0028	1.36	5700	0.0049	0.9961	0.9922	0.9941	0.9990
0.0034	1.43	6000	0.0048	0.9952	0.9937	0.9945	0.9990
0.0034	1.5	6300	0.0050	0.9936	0.9937	0.9937	0.9989
0.0022	1.57	6600	0.0046	0.9937	0.9934	0.9936	0.9989
0.0022	1.65	6900	0.0042	0.9954	0.9929	0.9941	0.9990
0.0039	1.72	7200	0.0042	0.9959	0.9931	0.9945	0.9990
0.003	1.79	7500	0.0039	0.9968	0.9927	0.9947	0.9991

Framework versions

Transformers 4.26.1
Pytorch 1.13.1+cu116
Datasets 2.9.0
Tokenizers 0.13.2

Task: Token Classification
Library: transformers
Type: camembert
Created: Feb 8, 2023
Updated: Feb 10, 2023