windowz_test-022625

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

• Accuracy: 0.9907
• F1: 0.9909
• Iou: 0.9830
• Per Class Metrics: {0: {'f1': 0.99734, 'iou': 0.9947, 'accuracy': 0.99602}, 1: {'f1': 0.9807, 'iou': 0.96214, 'accuracy': 0.99071}, 2: {'f1': 0.74699, 'iou': 0.59616, 'accuracy': 0.99465}}
• Loss: 0.0222

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 1000
• num_epochs: 100

Training results

Training Loss	Epoch	Step		Class Metrics	Validation Loss
0.486	5.0	12815	0.9750	{0: {'f1': 0.99477, 'iou': 0.9896, 'accuracy': 0.99218}, 1: {'f1': 0.97507, 'iou': 0.95135, 'accuracy': 0.98779}, 2: {'f1': 0.59112, 'iou': 0.41956, 'accuracy': 0.99417}}	0.1010
0.4424	10.0	25630	0.9841	{0: {'f1': 0.99787, 'iou': 0.99575, 'accuracy': 0.99682}, 1: {'f1': 0.98309, 'iou': 0.96675, 'accuracy': 0.99173}, 2: {'f1': 0.67276, 'iou': 0.50689, 'accuracy': 0.99485}}	0.0388
0.398	15.0	38445	0.9800	{0: {'f1': 0.99635, 'iou': 0.99272, 'accuracy': 0.99454}, 1: {'f1': 0.97804, 'iou': 0.95702, 'accuracy': 0.98935}, 2: {'f1': 0.71599, 'iou': 0.55762, 'accuracy': 0.99474}}	0.0339
0.3887	20.0	51260	0.9832	{0: {'f1': 0.99697, 'iou': 0.99395, 'accuracy': 0.99546}, 1: {'f1': 0.98169, 'iou': 0.96404, 'accuracy': 0.99117}, 2: {'f1': 0.76483, 'iou': 0.61921, 'accuracy': 0.99548}}	0.0228
0.3765	25.0	64075	0.9830	{0: {'f1': 0.99734, 'iou': 0.9947, 'accuracy': 0.99602}, 1: {'f1': 0.9807, 'iou': 0.96214, 'accuracy': 0.99071}, 2: {'f1': 0.74699, 'iou': 0.59616, 'accuracy': 0.99465}}	0.0222
0.4094	30.0	76890	0.9848	{0: {'f1': 0.99775, 'iou': 0.99551, 'accuracy': 0.99663}, 1: {'f1': 0.98255, 'iou': 0.9657, 'accuracy': 0.9916}, 2: {'f1': 0.7705, 'iou': 0.62667, 'accuracy': 0.99492}}	0.0345
0.371	35.0	89705	0.9836	{0: {'f1': 0.99757, 'iou': 0.99515, 'accuracy': 0.99636}, 1: {'f1': 0.98094, 'iou': 0.9626, 'accuracy': 0.99085}, 2: {'f1': 0.75391, 'iou': 0.60502, 'accuracy': 0.99445}}	0.0224
0.3752	40.0	102520	0.9826	{0: {'f1': 0.99777, 'iou': 0.99555, 'accuracy': 0.99666}, 1: {'f1': 0.97899, 'iou': 0.95885, 'accuracy': 0.98995}, 2: {'f1': 0.72023, 'iou': 0.56278, 'accuracy': 0.99326}}	0.0243

Framework versions

• Transformers 4.45.0
• Pytorch 2.5.1+cu124
• Datasets 2.21.0
• Tokenizers 0.20.3