riconoscimento_documenti_2

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	6	1.7741	0.1327
1.7625	2.0	12	1.4336	0.5174
1.7625	3.0	18	0.9371	0.8043
1.2151	4.0	24	0.3147	0.9973
0.3313	5.0	30	0.0508	1.0
0.3313	6.0	36	0.0072	1.0
0.0381	7.0	42	0.0051	1.0
0.0381	8.0	48	0.0054	0.9987
0.0204	9.0	54	0.0018	1.0
0.0303	10.0	60	0.0019	0.9987
0.0303	11.0	66	0.0018	0.9987
0.0164	12.0	72	0.0011	1.0
0.0164	13.0	78	0.0050	0.9987
0.0134	14.0	84	0.0039	0.9987
0.0194	15.0	90	0.0007	1.0
0.0194	16.0	96	0.0006	1.0
0.0053	17.0	102	0.0001	1.0
0.0053	18.0	108	0.0010	1.0
0.0055	19.0	114	0.0011	0.9987
0.0089	20.0	120	0.0007	1.0
0.0089	21.0	126	0.0001	1.0
0.0051	22.0	132	0.0001	1.0
0.0051	23.0	138	0.0019	1.0
0.0254	24.0	144	0.0001	1.0
0.005	25.0	150	0.0001	1.0
0.005	26.0	156	0.0000	1.0
0.0061	27.0	162	0.0002	1.0
0.0061	28.0	168	0.0001	1.0
0.0042	29.0	174	0.0001	1.0
0.0072	30.0	180	0.0006	1.0
0.0072	31.0	186	0.0001	1.0
0.0027	32.0	192	0.0001	1.0
0.0027	33.0	198	0.0001	1.0
0.0049	34.0	204	0.0001	1.0
0.0104	35.0	210	0.0005	1.0
0.0104	36.0	216	0.0002	1.0
0.0059	37.0	222	0.0001	1.0
0.0059	38.0	228	0.0002	1.0
0.0257	39.0	234	0.0000	1.0
0.0051	40.0	240	0.0000	1.0
0.0051	41.0	246	0.0000	1.0
0.0008	42.0	252	0.0000	1.0
0.0008	43.0	258	0.0000	1.0
0.0071	44.0	264	0.0000	1.0
0.0045	45.0	270	0.0000	1.0
0.0045	46.0	276	0.0000	1.0
0.0004	47.0	282	0.0000	1.0
0.0004	48.0	288	0.0000	1.0
0.0009	49.0	294	0.0000	1.0
0.0046	50.0	300	0.0000	1.0

Framework versions

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