ChristianMDahl/segFormer-b3-horizontal

This model is a fine-tuned version of nvidia/mit-b3 on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.1286
• Validation Loss: 0.1658
• Epoch: 19

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
0.2493	0.2166	0
0.2089	0.2028	1
0.1938	0.1947	2
0.1836	0.1909	3
0.1759	0.1842	4
0.1690	0.1823	5
0.1636	0.1778	6
0.1588	0.1753	7
0.1543	0.1787	8
0.1511	0.1732	9
0.1480	0.1729	10
0.1445	0.1703	11
0.1420	0.1689	12
0.1397	0.1681	13
0.1372	0.1669	14
0.1358	0.1700	15
0.1334	0.1698	16
0.1317	0.1662	17
0.1301	0.1683	18
0.1286	0.1658	19

Framework versions

• Transformers 4.28.1
• TensorFlow 2.10.1
• Datasets 2.12.0
• Tokenizers 0.13.0.dev0