line_detector_math
This model is a fine-tuned version of vikp/line_detector_2 on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1602
- Heatmap Mean Iou: 92.4902
- Heatmap Box Count: 40.9407
- Heatmap Correct Box Count: 41.4774
- Affinity Mean Iou: 23.9998
- Affinity Box Count: 218.5248
- Affinity Correct Box Count: 121.3200
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: 6
- eval_batch_size: 6
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 8
- total_train_batch_size: 48
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
Heatmap Mean Iou |
Heatmap Box Count |
Heatmap Correct Box Count |
Affinity Mean Iou |
Affinity Box Count |
Affinity Correct Box Count |
| 0.1625 |
2.72 |
1000 |
0.1602 |
92.4902 |
40.9407 |
41.4774 |
23.9998 |
218.5248 |
121.3200 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.2.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
