---
library_name: transformers
license: other
base_model: nvidia/mit-b0
tags:
- vision
- image-segmentation
- generated_from_trainer
model-index:
- name: segformer-b0-finetuned-batch1-30nov
  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. -->

# segformer-b0-finetuned-batch1-30nov

This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the PushkarA07/batch1-tiles dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0966
- Mean Iou: 0.5771
- Mean Accuracy: 0.6248
- Overall Accuracy: 0.9912
- Accuracy Abnormality: 0.2533
- Iou Abnormality: 0.1629

## 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: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Abnormality | Iou Abnormality |
|:-------------:|:------:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:--------------------:|:---------------:|
| 0.5472        | 0.4167 | 10   | 0.6708          | 0.4927   | 0.8372        | 0.9235           | 0.7497               | 0.0623          |
| 0.4938        | 0.8333 | 20   | 0.6279          | 0.5549   | 0.7553        | 0.9766           | 0.5310               | 0.1333          |
| 0.4212        | 1.25   | 30   | 0.5862          | 0.5301   | 0.7733        | 0.9634           | 0.5807               | 0.0970          |
| 0.3701        | 1.6667 | 40   | 0.4789          | 0.5446   | 0.6614        | 0.9811           | 0.3373               | 0.1080          |
| 0.2942        | 2.0833 | 50   | 0.4493          | 0.5046   | 0.7100        | 0.9499           | 0.4667               | 0.0594          |
| 0.3006        | 2.5    | 60   | 0.2510          | 0.5474   | 0.6485        | 0.9833           | 0.3091               | 0.1116          |
| 0.2309        | 2.9167 | 70   | 0.3163          | 0.5131   | 0.6837        | 0.9610           | 0.4027               | 0.0654          |
| 0.2249        | 3.3333 | 80   | 0.3296          | 0.5296   | 0.7012        | 0.9703           | 0.4284               | 0.0891          |
| 0.2643        | 3.75   | 90   | 0.3136          | 0.5328   | 0.7081        | 0.9713           | 0.4413               | 0.0943          |
| 0.2062        | 4.1667 | 100  | 0.2595          | 0.5395   | 0.6919        | 0.9761           | 0.4039               | 0.1029          |
| 0.1932        | 4.5833 | 110  | 0.2060          | 0.5752   | 0.6311        | 0.9905           | 0.2667               | 0.1599          |
| 0.1693        | 5.0    | 120  | 0.2871          | 0.5652   | 0.6707        | 0.9859           | 0.3512               | 0.1445          |
| 0.1555        | 5.4167 | 130  | 0.1601          | 0.5623   | 0.6411        | 0.9876           | 0.2898               | 0.1370          |
| 0.1344        | 5.8333 | 140  | 0.1512          | 0.5732   | 0.6543        | 0.9886           | 0.3155               | 0.1578          |
| 0.1531        | 6.25   | 150  | 0.1447          | 0.5706   | 0.6309        | 0.9899           | 0.2669               | 0.1514          |
| 0.149         | 6.6667 | 160  | 0.1763          | 0.5708   | 0.6415        | 0.9891           | 0.2891               | 0.1526          |
| 0.2171        | 7.0833 | 170  | 0.1883          | 0.5695   | 0.6407        | 0.9890           | 0.2876               | 0.1500          |
| 0.1207        | 7.5    | 180  | 0.1804          | 0.5725   | 0.6363        | 0.9897           | 0.2780               | 0.1552          |
| 0.106         | 7.9167 | 190  | 0.1413          | 0.5712   | 0.6299        | 0.9900           | 0.2648               | 0.1524          |
| 0.1234        | 8.3333 | 200  | 0.1407          | 0.5650   | 0.6124        | 0.9904           | 0.2292               | 0.1396          |
| 0.2295        | 8.75   | 210  | 0.0943          | 0.5734   | 0.6150        | 0.9914           | 0.2334               | 0.1555          |
| 0.1301        | 9.1667 | 220  | 0.0870          | 0.5746   | 0.6135        | 0.9917           | 0.2302               | 0.1575          |
| 0.1581        | 9.5833 | 230  | 0.0895          | 0.5747   | 0.6129        | 0.9917           | 0.2289               | 0.1577          |
| 0.1062        | 10.0   | 240  | 0.0966          | 0.5771   | 0.6248        | 0.9912           | 0.2533               | 0.1629          |


### Framework versions

- Transformers 4.46.2
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3