segformer-b0-scene-parse-150

This model is a fine-tuned version of DiTo97/binarization-segformer-b3 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1587
• Mean Iou: 0.5023
• Mean Accuracy: 0.5190
• Overall Accuracy: 0.9675
• Per Category Iou: [0.037137244396821145, 0.9674848565106324]
• Per Category Accuracy: [0.042339407512979745, 0.9957297061214373]

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: 6e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Per Category Iou	Per Category Accuracy
No log	1.0	133	1.3526	0.0201	0.5031	0.0397	[0.02975604398011791, 0.01052369675464703]	[0.9955767077267638, 0.010525115803790754]
No log	2.0	266	0.3525	0.4867	0.5015	0.9688	[0.004574905664611646, 0.9688262452629481]	[0.004842376734873935, 0.9982177109197249]
No log	3.0	399	0.2574	0.4889	0.5038	0.9688	[0.008923098533993973, 0.9688274858196525]	[0.00948454307916794, 0.9980818849540622]
0.6609	4.0	532	0.1598	0.4864	0.5012	0.9704	[0.0024469999746512434, 0.970368841777746]	[0.0024568190301672943, 0.9998776738732703]
0.6609	5.0	665	0.1452	0.4856	0.5004	0.9704	[0.000846027983221573, 0.9703602128333251]	[0.0008483491126268282, 0.9999163630668406]
0.6609	6.0	798	0.1447	0.4905	0.5053	0.9699	[0.011117650258900796, 0.9698808507647041]	[0.01144253283111066, 0.9991091658009873]
0.6609	7.0	931	0.1675	0.4998	0.5157	0.9684	[0.031259421511502704, 0.9683566316036148]	[0.03448030133360481, 0.9968589340278906]
0.4063	8.0	1064	0.1428	0.4930	0.5078	0.9697	[0.01628017129826309, 0.9696496379088535]	[0.016970375648987072, 0.9987075843853725]
0.4063	9.0	1197	0.1923	0.5086	0.5297	0.9641	[0.053108550932340105, 0.9640450905222431]	[0.0680308120397706, 0.9914344815058413]
0.4063	10.0	1330	0.1587	0.5023	0.5190	0.9675	[0.037137244396821145, 0.9674848565106324]	[0.042339407512979745, 0.9957297061214373]

Framework versions

• Transformers 4.37.0
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0