metadata
library_name: transformers
license: other
base_model: nvidia/mit-b0
tags:
- vision
- image-segmentation
- generated_from_trainer
model-index:
- name: earthquake
results: []
earthquake
This model is a fine-tuned version of nvidia/mit-b0 on the gokceKy/earthquake dataset. It achieves the following results on the evaluation set:
- Loss: 1.6395
- Mean Iou: 0.1763
- Mean Accuracy: 0.4255
- Overall Accuracy: 0.6211
- Accuracy Background: nan
- Accuracy Car: nan
- Accuracy Earthquake-roads: 0.4591
- Accuracy Other: 0.8175
- Accuracy Road: 0.0
- Accuracy Road-cracks: nan
- Accuracy Sky: nan
- Accuracy Wall: nan
- Iou Background: 0.0
- Iou Car: 0.0
- Iou Earthquake-roads: 0.4499
- Iou Other: 0.7841
- Iou Road: 0.0
- Iou Road-cracks: 0.0
- Iou Sky: 0.0
- Iou Wall: nan
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: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Use OptimizerNames.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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Background | Accuracy Car | Accuracy Earthquake-roads | Accuracy Other | Accuracy Road | Accuracy Road-cracks | Accuracy Sky | Accuracy Wall | Iou Background | Iou Car | Iou Earthquake-roads | Iou Other | Iou Road | Iou Road-cracks | Iou Sky | Iou Wall |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1.0736 | 2.5 | 5 | 1.6917 | 0.1818 | 0.3763 | 0.5984 | nan | nan | 0.3113 | 0.8176 | 0.0 | nan | nan | nan | 0.0 | 0.0 | 0.3077 | 0.7830 | 0.0 | 0.0 | nan | nan |
| 0.8949 | 5.0 | 10 | 1.6395 | 0.1763 | 0.4255 | 0.6211 | nan | nan | 0.4591 | 0.8175 | 0.0 | nan | nan | nan | 0.0 | 0.0 | 0.4499 | 0.7841 | 0.0 | 0.0 | 0.0 | nan |
Framework versions
- Transformers 4.46.2
- Pytorch 2.4.1+cpu
- Datasets 3.1.0
- Tokenizers 0.20.3