|
|
--- |
|
|
library_name: transformers |
|
|
license: apple-amlr |
|
|
base_model: apple/deeplabv3-mobilevit-small |
|
|
tags: |
|
|
- vision |
|
|
- image-segmentation |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: deeplabv3-mobilevit-small_corm |
|
|
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. --> |
|
|
|
|
|
# deeplabv3-mobilevit-small_corm |
|
|
|
|
|
This model is a fine-tuned version of [apple/deeplabv3-mobilevit-small](https://huggingface.co/apple/deeplabv3-mobilevit-small) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.7777 |
|
|
- Mean Iou: 0.4137 |
|
|
- Mean Accuracy: 0.5038 |
|
|
- Overall Accuracy: 0.7714 |
|
|
- Accuracy Background: 0.9998 |
|
|
- Accuracy Corm: 0.0748 |
|
|
- Accuracy Damage: 0.4368 |
|
|
- Iou Background: 0.7626 |
|
|
- Iou Corm: 0.0734 |
|
|
- Iou Damage: 0.4050 |
|
|
|
|
|
## 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: 4 |
|
|
- eval_batch_size: 4 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 5 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Background | Accuracy Corm | Accuracy Damage | Iou Background | Iou Corm | Iou Damage | |
|
|
|:-------------:|:------:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:-------------------:|:-------------:|:---------------:|:--------------:|:--------:|:----------:| |
|
|
| 1.0611 | 0.3077 | 20 | 1.0627 | 0.4251 | 0.6398 | 0.6594 | 0.6709 | 0.5632 | 0.6853 | 0.6508 | 0.2723 | 0.3522 | |
|
|
| 1.0077 | 0.6154 | 40 | 1.0227 | 0.5359 | 0.6715 | 0.8018 | 0.9058 | 0.4028 | 0.7060 | 0.8285 | 0.3049 | 0.4743 | |
|
|
| 1.0089 | 0.9231 | 60 | 0.9859 | 0.5131 | 0.6138 | 0.8167 | 0.9804 | 0.2104 | 0.6506 | 0.8345 | 0.1883 | 0.5167 | |
|
|
| 0.9397 | 1.2308 | 80 | 0.9637 | 0.4767 | 0.5831 | 0.8103 | 0.9867 | 0.0722 | 0.6904 | 0.8293 | 0.0701 | 0.5307 | |
|
|
| 0.9347 | 1.5385 | 100 | 0.9257 | 0.4544 | 0.5545 | 0.7991 | 0.9946 | 0.0517 | 0.6172 | 0.8103 | 0.0504 | 0.5024 | |
|
|
| 0.9007 | 1.8462 | 120 | 0.9054 | 0.4458 | 0.5428 | 0.7926 | 0.9968 | 0.0678 | 0.5637 | 0.8012 | 0.0658 | 0.4705 | |
|
|
| 0.8787 | 2.1538 | 140 | 0.8756 | 0.4195 | 0.5140 | 0.7780 | 0.9986 | 0.0506 | 0.4927 | 0.7790 | 0.0495 | 0.4301 | |
|
|
| 0.8757 | 2.4615 | 160 | 0.8501 | 0.4035 | 0.4967 | 0.7677 | 0.9994 | 0.0659 | 0.4247 | 0.7656 | 0.0645 | 0.3804 | |
|
|
| 0.841 | 2.7692 | 180 | 0.8339 | 0.4199 | 0.5148 | 0.7799 | 0.9987 | 0.0283 | 0.5174 | 0.7791 | 0.0279 | 0.4528 | |
|
|
| 0.8268 | 3.0769 | 200 | 0.8246 | 0.4358 | 0.5279 | 0.7844 | 0.9989 | 0.0809 | 0.5040 | 0.7826 | 0.0789 | 0.4460 | |
|
|
| 0.8306 | 3.3846 | 220 | 0.8095 | 0.4034 | 0.4968 | 0.7690 | 0.9995 | 0.0461 | 0.4448 | 0.7653 | 0.0455 | 0.3995 | |
|
|
| 0.826 | 3.6923 | 240 | 0.7928 | 0.4174 | 0.5078 | 0.7731 | 0.9997 | 0.0846 | 0.4391 | 0.7663 | 0.0826 | 0.4034 | |
|
|
| 0.7873 | 4.0 | 260 | 0.7915 | 0.4150 | 0.5046 | 0.7713 | 0.9996 | 0.0842 | 0.4299 | 0.7616 | 0.0824 | 0.4009 | |
|
|
| 0.8031 | 4.3077 | 280 | 0.7805 | 0.4022 | 0.4928 | 0.7648 | 0.9998 | 0.0826 | 0.3960 | 0.7557 | 0.0811 | 0.3699 | |
|
|
| 0.7881 | 4.6154 | 300 | 0.7791 | 0.4352 | 0.5265 | 0.7848 | 0.9995 | 0.0659 | 0.5142 | 0.7808 | 0.0645 | 0.4604 | |
|
|
| 0.7883 | 4.9231 | 320 | 0.7777 | 0.4137 | 0.5038 | 0.7714 | 0.9998 | 0.0748 | 0.4368 | 0.7626 | 0.0734 | 0.4050 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.44.1 |
|
|
- Pytorch 2.6.0+cpu |
|
|
- Datasets 2.21.0 |
|
|
- Tokenizers 0.19.1 |
|
|
|
