|
--- |
|
license: apache-2.0 |
|
base_model: distilbert-base-cased |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- f1 |
|
model-index: |
|
- name: persuasive_essays_distilbert_cased |
|
results: [] |
|
language: |
|
- en |
|
--- |
|
|
|
# persuasive_essays_distilbert_cased |
|
|
|
## Model description |
|
|
|
This model is a fine-tuned version of [distilbert-base-cased](https://huggingface.co/distilbert-base-cased) on the [emnlp2017-claim-identification/persuasive_essays](https://github.com/UKPLab/emnlp2017-claim-identification) dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4249 |
|
- Accuracy: 0.8101 |
|
- Macro F1: 0.7662 |
|
- Claim F1: 0.665 |
|
|
|
## Intended uses & limitations |
|
|
|
Text classification for claims on full sentences. The model perfoms better at in-domain classification. Cross-domain classification is severely limited. |
|
|
|
## Training and evaluation data |
|
|
|
Based on [Stab and Gurevych (2017)](https://aclanthology.org/J17-3005.pdf) persuasive essays corpus, preprocessed by [Daxenberger et al. (2017)]((https://github.com/UKPLab/emnlp2017-claim-identification). |
|
|
|
Original dataset |
|
- docs: 402 |
|
- tokens: 147,271 |
|
- total instances: 7,116 (65 duplicates) |
|
- #claims: 2,108 (29.62%) |
|
|
|
Trimmed datast used for training |
|
- total instances: **7051** (65 duplicates removed) |
|
- #claims: **2093** (29.68%) |
|
- train/test split: 80/20, stratified |
|
|
|
## 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: 2 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Macro F1 | Claim F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:| |
|
| No log | 1.0 | 353 | 0.4369 | 0.7931 | 0.7574 | 0.6644 | |
|
| 0.4492 | 2.0 | 706 | 0.4249 | 0.8101 | 0.7662 | 0.665 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.37.2 |
|
- Pytorch 2.2.0 |
|
- Datasets 2.17.0 |
|
- Tokenizers 0.15.2 |
