riconoscimento_documenti

This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0000
• Accuracy: 1.0

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	1	1.9560	0.0
No log	2.0	3	1.4216	0.7375
No log	3.0	5	0.6008	1.0
No log	4.0	6	0.2696	1.0
No log	5.0	7	0.0996	1.0
No log	6.0	9	0.0089	1.0
0.4721	7.0	11	0.0011	1.0
0.4721	8.0	12	0.0005	1.0
0.4721	9.0	13	0.0002	1.0
0.4721	10.0	15	0.0001	1.0
0.4721	11.0	17	0.0000	1.0
0.4721	12.0	18	0.0000	1.0
0.4721	13.0	19	0.0000	1.0
0.0003	14.0	21	0.0000	1.0
0.0003	15.0	23	0.0000	1.0
0.0003	16.0	24	0.0000	1.0
0.0003	17.0	25	0.0000	1.0
0.0003	18.0	27	0.0000	1.0
0.0003	19.0	29	0.0000	1.0
0.0	20.0	30	0.0000	1.0
0.0	21.0	31	0.0000	1.0
0.0	22.0	33	0.0000	1.0
0.0	23.0	35	0.0000	1.0
0.0	24.0	36	0.0000	1.0
0.0	25.0	37	0.0000	1.0
0.0	26.0	39	0.0000	1.0
0.0	27.0	41	0.0000	1.0
0.0	28.0	42	0.0000	1.0
0.0	29.0	43	0.0000	1.0
0.0	30.0	45	0.0000	1.0
0.0	31.0	47	0.0000	1.0
0.0	32.0	48	0.0000	1.0
0.0	33.0	49	0.0000	1.0
0.0	33.33	50	0.0000	1.0

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2