Omar280X/personal_license_segmentation

This model is a fine-tuned version of nvidia/segformer-b0-finetuned-ade-512-512 on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.0104
• Validation Loss: 0.0105
• Epoch: 49

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:

• optimizer: {'name': 'Adam', 'learning_rate': 6e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
• training_precision: float32

Training results

Train Loss	Validation Loss	Epoch
0.0804	0.0281	0
0.0291	0.0244	1
0.0272	0.0234	2
0.0240	0.0222	3
0.0224	0.0212	4
0.0214	0.0201	5
0.0224	0.0199	6
0.0200	0.0189	7
0.0210	0.0187	8
0.0196	0.0184	9
0.0185	0.0174	10
0.0180	0.0169	11
0.0178	0.0165	12
0.0167	0.0160	13
0.0162	0.0155	14
0.0158	0.0147	15
0.0151	0.0143	16
0.0147	0.0138	17
0.0146	0.0137	18
0.0144	0.0134	19
0.0143	0.0131	20
0.0134	0.0130	21
0.0132	0.0132	22
0.0130	0.0132	23
0.0130	0.0124	24
0.0127	0.0122	25
0.0127	0.0122	26
0.0124	0.0119	27
0.0122	0.0123	28
0.0123	0.0117	29
0.0118	0.0115	30
0.0117	0.0117	31
0.0116	0.0115	32
0.0115	0.0117	33
0.0114	0.0113	34
0.0113	0.0111	35
0.0110	0.0109	36
0.0125	0.0109	37
0.0114	0.0108	38
0.0107	0.0108	39
0.0107	0.0108	40
0.0122	0.0112	41
0.0109	0.0108	42
0.0106	0.0106	43
0.0105	0.0107	44
0.0106	0.0106	45
0.0105	0.0108	46
0.0106	0.0110	47
0.0103	0.0108	48
0.0104	0.0105	49

Framework versions

• Transformers 4.37.0
• TensorFlow 2.5.0
• Tokenizers 0.15.2