File size: 3,738 Bytes
38fcc13 c39c26f 38fcc13 c39c26f 38fcc13 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
---
library_name: transformers
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: windowz_test-022625
results: []
---
<!-- 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. -->
# windowz_test-022625
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
It achieves the following results on the evaluation set:
- Accuracy: 0.9907
- F1: 0.9909
- Iou: 0.9830
- Per Class Metrics: {0: {'f1': 0.99734, 'iou': 0.9947, 'accuracy': 0.99602}, 1: {'f1': 0.9807, 'iou': 0.96214, 'accuracy': 0.99071}, 2: {'f1': 0.74699, 'iou': 0.59616, 'accuracy': 0.99465}}
- Loss: 0.0222
## 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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 100
### Training results
| Training Loss | Epoch | Step | | Class Metrics | Validation Loss |
|:-------------:|:-----:|:------:|:------:|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|:---------------:|
| 0.486 | 5.0 | 12815 | 0.9750 | {0: {'f1': 0.99477, 'iou': 0.9896, 'accuracy': 0.99218}, 1: {'f1': 0.97507, 'iou': 0.95135, 'accuracy': 0.98779}, 2: {'f1': 0.59112, 'iou': 0.41956, 'accuracy': 0.99417}} | 0.1010 |
| 0.4424 | 10.0 | 25630 | 0.9841 | {0: {'f1': 0.99787, 'iou': 0.99575, 'accuracy': 0.99682}, 1: {'f1': 0.98309, 'iou': 0.96675, 'accuracy': 0.99173}, 2: {'f1': 0.67276, 'iou': 0.50689, 'accuracy': 0.99485}} | 0.0388 |
| 0.398 | 15.0 | 38445 | 0.9800 | {0: {'f1': 0.99635, 'iou': 0.99272, 'accuracy': 0.99454}, 1: {'f1': 0.97804, 'iou': 0.95702, 'accuracy': 0.98935}, 2: {'f1': 0.71599, 'iou': 0.55762, 'accuracy': 0.99474}} | 0.0339 |
| 0.3887 | 20.0 | 51260 | 0.9832 | {0: {'f1': 0.99697, 'iou': 0.99395, 'accuracy': 0.99546}, 1: {'f1': 0.98169, 'iou': 0.96404, 'accuracy': 0.99117}, 2: {'f1': 0.76483, 'iou': 0.61921, 'accuracy': 0.99548}} | 0.0228 |
| 0.3765 | 25.0 | 64075 | 0.9830 | {0: {'f1': 0.99734, 'iou': 0.9947, 'accuracy': 0.99602}, 1: {'f1': 0.9807, 'iou': 0.96214, 'accuracy': 0.99071}, 2: {'f1': 0.74699, 'iou': 0.59616, 'accuracy': 0.99465}} | 0.0222 |
| 0.4094 | 30.0 | 76890 | 0.9848 | {0: {'f1': 0.99775, 'iou': 0.99551, 'accuracy': 0.99663}, 1: {'f1': 0.98255, 'iou': 0.9657, 'accuracy': 0.9916}, 2: {'f1': 0.7705, 'iou': 0.62667, 'accuracy': 0.99492}} | 0.0345 |
| 0.371 | 35.0 | 89705 | 0.9836 | {0: {'f1': 0.99757, 'iou': 0.99515, 'accuracy': 0.99636}, 1: {'f1': 0.98094, 'iou': 0.9626, 'accuracy': 0.99085}, 2: {'f1': 0.75391, 'iou': 0.60502, 'accuracy': 0.99445}} | 0.0224 |
| 0.3752 | 40.0 | 102520 | 0.9826 | {0: {'f1': 0.99777, 'iou': 0.99555, 'accuracy': 0.99666}, 1: {'f1': 0.97899, 'iou': 0.95885, 'accuracy': 0.98995}, 2: {'f1': 0.72023, 'iou': 0.56278, 'accuracy': 0.99326}} | 0.0243 |
### Framework versions
- Transformers 4.45.0
- Pytorch 2.5.1+cu124
- Datasets 2.21.0
- Tokenizers 0.20.3
|