metadata
license: mit
base_model: microsoft/deberta-v3-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: unga-climate-classifier
results: []
ECCA climate classifier
This model is a fine-tuned version of microsoft/deberta-v3-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0936
- Accuracy: 0.9798
- F1 Macro: 0.9765
- Accuracy Balanced: 0.9751
- F1 Micro: 0.9798
- Precision Macro: 0.9780
- Recall Macro: 0.9751
- Precision Micro: 0.9798
- Recall Micro: 0.9798
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 80
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Accuracy Balanced | F1 Micro | Precision Macro | Recall Macro | Precision Micro | Recall Micro |
|---|---|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 123 | 0.1778 | 0.9609 | 0.9543 | 0.9510 | 0.9609 | 0.9577 | 0.9510 | 0.9609 | 0.9609 |
| No log | 2.0 | 246 | 0.1614 | 0.9680 | 0.9626 | 0.9593 | 0.9680 | 0.9661 | 0.9593 | 0.9680 | 0.9680 |
| No log | 3.0 | 369 | 0.1598 | 0.9680 | 0.9626 | 0.9593 | 0.9680 | 0.9661 | 0.9593 | 0.9680 | 0.9680 |
| No log | 4.0 | 492 | 0.1191 | 0.9703 | 0.9653 | 0.9610 | 0.9703 | 0.9699 | 0.9610 | 0.9703 | 0.9703 |
| 0.1357 | 5.0 | 615 | 0.1400 | 0.9727 | 0.9681 | 0.9638 | 0.9727 | 0.9727 | 0.9638 | 0.9727 | 0.9727 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.6.0
- Tokenizers 0.15.1