DataIntelligenceTeam/perioli_manifesti_v5.6 - AI Model Zoo

perioli_manifesti_v5.6

This model is a fine-tuned version of microsoft/layoutlmv3-base on the sroie dataset. It achieves the following results on the evaluation set:

Loss: 0.0271
Precision: 0.9305
Recall: 0.9623
F1: 0.9461
Accuracy: 0.9934

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: 3500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.47	100	0.1846	0.6553	0.7448	0.6972	0.9597
No log	0.93	200	0.0547	0.8780	0.9182	0.8977	0.9865
No log	1.4	300	0.0442	0.9007	0.9371	0.9185	0.9895
No log	1.87	400	0.0446	0.8823	0.9362	0.9085	0.9886
0.1459	2.34	500	0.0360	0.8995	0.9488	0.9235	0.9908
0.1459	2.8	600	0.0315	0.9205	0.9569	0.9383	0.9923
0.1459	3.27	700	0.0309	0.9316	0.9668	0.9489	0.9937
0.1459	3.74	800	0.0259	0.9121	0.9506	0.9309	0.9918
0.1459	4.21	900	0.0267	0.9264	0.9614	0.9436	0.9941
0.0247	4.67	1000	0.0272	0.9282	0.9641	0.9458	0.9937
0.0247	5.14	1100	0.0221	0.9355	0.9641	0.9496	0.9947
0.0247	5.61	1200	0.0236	0.9494	0.9614	0.9554	0.9946
0.0247	6.07	1300	0.0331	0.9050	0.9416	0.9229	0.9898
0.0247	6.54	1400	0.0199	0.9278	0.9587	0.9430	0.9936
0.016	7.01	1500	0.0180	0.9465	0.9533	0.9499	0.9947
0.016	7.48	1600	0.0236	0.9437	0.9641	0.9538	0.9951
0.016	7.94	1700	0.0220	0.9359	0.9704	0.9528	0.9946
0.016	8.41	1800	0.0271	0.9185	0.9623	0.9399	0.9923
0.016	8.88	1900	0.0315	0.9294	0.9695	0.9490	0.9936
0.011	9.35	2000	0.0305	0.9186	0.9632	0.9404	0.9926
0.011	9.81	2100	0.0286	0.9271	0.9596	0.9430	0.9930
0.011	10.28	2200	0.0242	0.9296	0.9497	0.9396	0.9928
0.011	10.75	2300	0.0316	0.9183	0.9488	0.9333	0.9908
0.011	11.21	2400	0.0293	0.9311	0.9596	0.9451	0.9932
0.0069	11.68	2500	0.0285	0.9297	0.9632	0.9462	0.9932
0.0069	12.15	2600	0.0309	0.9114	0.9524	0.9315	0.9918
0.0069	12.62	2700	0.0290	0.9206	0.9587	0.9393	0.9928
0.0069	13.08	2800	0.0270	0.9323	0.9650	0.9483	0.9937
0.0069	13.55	2900	0.0249	0.9363	0.9641	0.9500	0.9941
0.0059	14.02	3000	0.0273	0.9297	0.9632	0.9462	0.9933
0.0059	14.49	3100	0.0271	0.9278	0.9587	0.9430	0.9930
0.0059	14.95	3200	0.0297	0.9248	0.9614	0.9427	0.9928
0.0059	15.42	3300	0.0275	0.9305	0.9623	0.9461	0.9933
0.0059	15.89	3400	0.0275	0.9296	0.9605	0.9448	0.9933
0.0047	16.36	3500	0.0271	0.9305	0.9623	0.9461	0.9934

Framework versions

Transformers 4.28.0
Pytorch 2.0.1+cu118
Datasets 2.2.2
Tokenizers 0.13.3