segformer-b0-finetuned-pokemon

This model is a fine-tuned version of ydmeira/segformer-b0-finetuned-pokemon on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0157
• Mean Iou: 0.4970
• Mean Accuracy: 0.9940
• Overall Accuracy: 0.9940
• Per Category Iou: [0.0, 0.9940101727137823]
• Per Category Accuracy: [nan, 0.9940101727137823]

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: 24
• eval_batch_size: 24
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Per Category Iou	Per Category Accuracy
0.0175	45.0	1305	0.0157	0.4971	0.9943	0.9943	[0.0, 0.9942906494536522]	[nan, 0.9942906494536522]
0.018	46.0	1334	0.0157	0.4968	0.9936	0.9936	[0.0, 0.9936369941650801]	[nan, 0.9936369941650801]
0.0185	47.0	1363	0.0157	0.4971	0.9943	0.9943	[0.0, 0.9942791789145462]	[nan, 0.9942791789145462]
0.018	48.0	1392	0.0157	0.4969	0.9937	0.9937	[0.0, 0.9937245121725857]	[nan, 0.9937245121725857]
0.0183	49.0	1421	0.0157	0.4969	0.9939	0.9939	[0.0, 0.9938530594161242]	[nan, 0.9938530594161242]
0.0196	50.0	1450	0.0157	0.4970	0.9940	0.9940	[0.0, 0.9940101727137823]	[nan, 0.9940101727137823]

Framework versions

• Transformers 4.20.1
• Pytorch 1.12.0+cu113
• Datasets 2.3.2
• Tokenizers 0.12.1