README.md

---
license: other
base_model: nvidia/segformer-b1-finetuned-ade-512-512
tags:
- vision
- image-segmentation
- generated_from_trainer
metrics:
- precision
model-index:
- name: segformer-b1-finetuned-segments-pv_v1_normalized_p100_4batch_try
 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. -->

[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/mouadn773/huggingface/runs/b36iwo31)
# segformer-b1-finetuned-segments-pv_v1_normalized_p100_4batch_try

This model is a fine-tuned version of [nvidia/segformer-b1-finetuned-ade-512-512](https://huggingface.co/nvidia/segformer-b1-finetuned-ade-512-512) on the mouadenna/satellite_PV_dataset_train_test_v1 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0012
- Mean Iou: 0.9586
- Precision: 0.9787

## 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: 0.0004
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 40
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Precision |
|:-------------:|:-------:|:----:|:---------------:|:--------:|:---------:|
| 0.2001 | 0.9989 | 229 | 0.0557 | 0.7433 | 0.8099 |
| 0.0234 | 1.9978 | 458 | 0.0111 | 0.8026 | 0.8735 |
| 0.0109 | 2.9967 | 687 | 0.0069 | 0.8264 | 0.9062 |
| 0.0077 | 4.0 | 917 | 0.0052 | 0.8659 | 0.9087 |
| 0.0053 | 4.9989 | 1146 | 0.0142 | 0.6652 | 0.6845 |
| 0.0055 | 5.9978 | 1375 | 0.0041 | 0.8885 | 0.9319 |
| 0.0041 | 6.9967 | 1604 | 0.0038 | 0.8855 | 0.9516 |
| 0.0046 | 8.0 | 1834 | 0.0041 | 0.8787 | 0.9430 |
| 0.0041 | 8.9989 | 2063 | 0.0035 | 0.8902 | 0.9105 |
| 0.0035 | 9.9978 | 2292 | 0.0028 | 0.9107 | 0.9475 |
| 0.0034 | 10.9967 | 2521 | 0.0027 | 0.9116 | 0.9390 |
| 0.0038 | 12.0 | 2751 | 0.0031 | 0.9001 | 0.9306 |
| 0.0033 | 12.9989 | 2980 | 0.0025 | 0.9160 | 0.9563 |
| 0.0029 | 13.9978 | 3209 | 0.0026 | 0.9153 | 0.9444 |
| 0.0026 | 14.9967 | 3438 | 0.0023 | 0.9218 | 0.9499 |
| 0.0027 | 16.0 | 3668 | 0.0027 | 0.9086 | 0.9525 |
| 0.0032 | 16.9989 | 3897 | 0.0029 | 0.9076 | 0.9439 |
| 0.0033 | 17.9978 | 4126 | 0.0024 | 0.9192 | 0.9459 |
| 0.0025 | 18.9967 | 4355 | 0.0027 | 0.9108 | 0.9538 |
| 0.0024 | 20.0 | 4585 | 0.0021 | 0.9276 | 0.9559 |
| 0.0022 | 20.9989 | 4814 | 0.0020 | 0.9316 | 0.9649 |
| 0.0021 | 21.9978 | 5043 | 0.0021 | 0.9287 | 0.9571 |
| 0.0025 | 22.9967 | 5272 | 0.0023 | 0.9217 | 0.9511 |
| 0.0022 | 24.0 | 5502 | 0.0020 | 0.9309 | 0.9676 |
| 0.002 | 24.9989 | 5731 | 0.0018 | 0.9360 | 0.9613 |
| 0.002 | 25.9978 | 5960 | 0.0017 | 0.9394 | 0.9621 |
| 0.0019 | 26.9967 | 6189 | 0.0017 | 0.9403 | 0.9664 |
| 0.0018 | 28.0 | 6419 | 0.0017 | 0.9405 | 0.9566 |
| 0.0017 | 28.9989 | 6648 | 0.0016 | 0.9438 | 0.9695 |
| 0.0017 | 29.9978 | 6877 | 0.0015 | 0.9469 | 0.9755 |
| 0.002 | 30.9967 | 7106 | 0.0016 | 0.9448 | 0.9672 |
| 0.0018 | 32.0 | 7336 | 0.0015 | 0.9459 | 0.9693 |
| 0.0016 | 32.9989 | 7565 | 0.0014 | 0.9486 | 0.9674 |
| 0.0015 | 33.9978 | 7794 | 0.0013 | 0.9528 | 0.9761 |
| 0.0015 | 34.9967 | 8023 | 0.0013 | 0.9520 | 0.9732 |
| 0.0014 | 36.0 | 8253 | 0.0013 | 0.9541 | 0.9705 |
| 0.0014 | 36.9989 | 8482 | 0.0012 | 0.9563 | 0.9739 |
| 0.0014 | 37.9978 | 8711 | 0.0012 | 0.9575 | 0.9764 |
| 0.0014 | 38.9967 | 8940 | 0.0012 | 0.9581 | 0.9766 |
| 0.0014 | 39.9564 | 9160 | 0.0012 | 0.9586 | 0.9787 |


### Framework versions

- Transformers 4.42.3
- Pytorch 2.1.2
- Datasets 2.20.0
- Tokenizers 0.19.1