<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
segformer-roof
This model is a fine-tuned version of nvidia/mit-b4 on the None dataset. It achieves the following results on the evaluation set:
- eval_loss: 0.1252
- eval_mean_iou: 0.6127
- eval_mean_accuracy: 0.6824
- eval_overall_accuracy: 0.9576
- eval_per_category_iou: [0.9565401111001023, 0.5056775417338342, 0.6199915247210127, 0.3684174133417381]
- eval_per_category_accuracy: [0.9868643557839373, 0.6055775784407109, 0.6926147942226376, 0.44438043699630503]
- eval_runtime: 584.3554
- eval_samples_per_second: 1.415
- eval_steps_per_second: 0.354
- epoch: 18.0
- step: 16740
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: 8
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3