metadata
license: other
tags:
- vision
- image-segmentation
- generated_from_trainer
base_model: nvidia/mit-b0
model-index:
- name: segformer-b0-finetuned-sudoku
results: []
segformer-b0-finetuned-sudoku
This model is a fine-tuned version of nvidia/mit-b0 on the mrkprc1/SudokuBoundaries2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.6963
- Mean Iou: 0.2715
- Mean Accuracy: 0.4994
- Overall Accuracy: 0.4914
- Accuracy Unlabelled: 0.9207
- Accuracy Sudoku-boundary: 0.0782
- Iou Unlabelled: 0.4703
- Iou Sudoku-boundary: 0.0726
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-07
- train_batch_size: 2
- eval_batch_size: 2
- 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 | Accuracy Unlabelled | Accuracy Sudoku-boundary | Iou Unlabelled | Iou Sudoku-boundary |
---|---|---|---|---|---|---|---|---|---|---|
0.5345 | 3.33 | 20 | 0.6911 | 0.3154 | 0.5281 | 0.5355 | 0.1414 | 0.9147 | 0.1299 | 0.5008 |
1.1081 | 6.67 | 40 | 0.6904 | 0.3388 | 0.5351 | 0.5411 | 0.2242 | 0.8460 | 0.1933 | 0.4844 |
0.3015 | 10.0 | 60 | 0.6963 | 0.2715 | 0.4994 | 0.4914 | 0.9207 | 0.0782 | 0.4703 | 0.0726 |
Framework versions
- Transformers 4.37.1
- Pytorch 2.1.2
- Datasets 2.16.1
- Tokenizers 0.15.1