<!-- 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. -->
cc_narratives_robertamodel3
This model is a fine-tuned version of nnisbett/cc-narratives_robertamodel2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1325
- F1: 0.8278
Model description
This model classifies climate-related sentences into either normative, delay, or economic based on the argument they express.
Intended uses & limitations
More information needed
Training and evaluation data
This model was trained on transcripts of interviews with UK members of parliament.
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.1795 | 1.0 | 22 | 0.5493 | 0.8459 |
| 0.1221 | 2.0 | 44 | 0.8102 | 0.8392 |
| 0.0821 | 3.0 | 66 | 0.6461 | 0.8682 |
| 0.0581 | 4.0 | 88 | 0.5884 | 0.8610 |
| 0.0352 | 5.0 | 110 | 0.6353 | 0.8878 |
| 0.0332 | 6.0 | 132 | 0.5653 | 0.9135 |
| 0.0176 | 7.0 | 154 | 0.5591 | 0.9051 |
| 0.0251 | 8.0 | 176 | 0.6146 | 0.8878 |
| 0.0152 | 9.0 | 198 | 0.5605 | 0.8963 |
| 0.0059 | 10.0 | 220 | 0.5578 | 0.8963 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
