---
library_name: transformers
tags:
- generated_from_trainer
model-index:
- name: glacier_segmentation_transformer
  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. -->

# glacier_segmentation_transformer

This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0488
- Mean Iou: 0.9289
- Mean Accuracy: 0.9613
- Overall Accuracy: 0.9689
- Per Category Iou: [0.9484225100312303, 0.875795429449281, 0.9626254685949275]
- Per Category Accuracy: [0.9725549195530643, 0.9263729934338871, 0.9850380570292713]

## 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: 0.00018
- train_batch_size: 100
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Per Category Iou                                            | Per Category Accuracy                                        |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:-----------------------------------------------------------:|:------------------------------------------------------------:|
| 0.1259        | 1.0   | 1405 | 0.0488          | 0.9289   | 0.9613        | 0.9689           | [0.9484225100312303, 0.875795429449281, 0.9626254685949275] | [0.9725549195530643, 0.9263729934338871, 0.9850380570292713] |


### Framework versions

- Transformers 4.45.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.0