layoutlmv3-finetuned-passport

EphronM/layoutlmv3-finetuned-passport

layoutlmv3-finetuned-passport is a token classification model from EphronM. Use it when you need labels on individual words, such as names. The card lists the license as cc-by-nc-sa-4.0.

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

Downloads · 30 days

1% of all-time downloads

All-time downloads

1.3K

Public

Parameters

126M

4 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors504 MB · 99%

At a glance

Task: Token Classification
License: cc-by-nc-sa-4.0
Model type: layoutlmv3
Access: Public
Created: Aug 6, 2024
Updated: Aug 7, 2024
SHA: 614faa9c

Try a prompt

Base models

microsoft/layoutlmv3-base

Datasets

EphronM/Annotated_passport_images

Model card

From the Hugging Face model README

layoutlmv3-finetuned-passport

This model is a fine-tuned version of microsoft/layoutlmv3-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0655
Precision: 0.9735
Recall: 0.9847
F1: 0.9790
Accuracy: 0.9892

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: 1e-05
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: 4000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	3.4483	100	0.6286	0.7930	0.7778	0.7853	0.8898
No log	6.8966	200	0.1945	0.9423	0.9387	0.9405	0.9719
No log	10.3448	300	0.0832	0.9730	0.9655	0.9692	0.9870
No log	13.7931	400	0.0558	0.9660	0.9808	0.9734	0.9870
0.398	17.2414	500	0.0524	0.9624	0.9808	0.9715	0.9870
0.398	20.6897	600	0.0462	0.9735	0.9847	0.9790	0.9892
0.398	24.1379	700	0.0543	0.9660	0.9808	0.9734	0.9870
0.398	27.5862	800	0.0463	0.9735	0.9847	0.9790	0.9892
0.398	31.0345	900	0.0569	0.9624	0.9808	0.9715	0.9870
0.0139	34.4828	1000	0.0729	0.9515	0.9770	0.9641	0.9849
0.0139	37.9310	1100	0.0656	0.9624	0.9808	0.9715	0.9870
0.0139	41.3793	1200	0.0609	0.9624	0.9808	0.9715	0.9870
0.0139	44.8276	1300	0.0525	0.9732	0.9732	0.9732	0.9892
0.0139	48.2759	1400	0.0735	0.9515	0.9770	0.9641	0.9849
0.0072	51.7241	1500	0.0491	0.9547	0.9693	0.9620	0.9870
0.0072	55.1724	1600	0.0416	0.9773	0.9885	0.9829	0.9914
0.0072	58.6207	1700	0.0472	0.9773	0.9885	0.9829	0.9914
0.0072	62.0690	1800	0.0543	0.9735	0.9847	0.9790	0.9892
0.0072	65.5172	1900	0.0619	0.9662	0.9847	0.9753	0.9892
0.0029	68.9655	2000	0.0670	0.9624	0.9808	0.9715	0.9870
0.0029	72.4138	2100	0.0770	0.9624	0.9808	0.9715	0.9870
0.0029	75.8621	2200	0.0700	0.9624	0.9808	0.9715	0.9870
0.0029	79.3103	2300	0.0655	0.9624	0.9808	0.9715	0.9870
0.0029	82.7586	2400	0.0684	0.9624	0.9808	0.9715	0.9870
0.0012	86.2069	2500	0.0700	0.9624	0.9808	0.9715	0.9870
0.0012	89.6552	2600	0.0696	0.9624	0.9808	0.9715	0.9870
0.0012	93.1034	2700	0.0619	0.9735	0.9847	0.9790	0.9892
0.0012	96.5517	2800	0.0630	0.9735	0.9847	0.9790	0.9892
0.0012	100.0	2900	0.0703	0.9733	0.9770	0.9751	0.9892
0.0009	103.4483	3000	0.0655	0.9735	0.9847	0.9790	0.9892
0.0009	106.8966	3100	0.0653	0.9735	0.9847	0.9790	0.9892
0.0009	110.3448	3200	0.0657	0.9735	0.9847	0.9790	0.9892
0.0009	113.7931	3300	0.0660	0.9735	0.9847	0.9790	0.9892
0.0009	117.2414	3400	0.0655	0.9735	0.9847	0.9790	0.9892
0.0008	120.6897	3500	0.0663	0.9735	0.9847	0.9790	0.9892
0.0008	124.1379	3600	0.0663	0.9735	0.9847	0.9790	0.9892
0.0008	127.5862	3700	0.0666	0.9735	0.9847	0.9790	0.9892
0.0008	131.0345	3800	0.0648	0.9735	0.9847	0.9790	0.9892
0.0008	134.4828	3900	0.0660	0.9735	0.9847	0.9790	0.9892
0.0009	137.9310	4000	0.0660	0.9735	0.9847	0.9790	0.9892

Framework versions

Transformers 4.44.0.dev0
Pytorch 2.1.0+cu121
Datasets 2.20.0
Tokenizers 0.19.1

Task: Token Classification
Type: layoutlmv3
License: cc-by-nc-sa-4.0
Languages: en
Created: Aug 6, 2024
Updated: Aug 7, 2024