poltextlab's picture
model card init
5e4f020 verified
---
---
language:
- nl
tags:
- text-classification
- pytorch
metrics:
- accuracy
- f1-score
---
# xlm-roberta-large-dutch-cap-v3
## Model description
An `xlm-roberta-large` model fine-tuned on dutch training data labeled with [major topic codes](https://www.comparativeagendas.net/pages/master-codebook) from the [Comparative Agendas Project](https://www.comparativeagendas.net/).
## How to use the model
This snippet prints the three most probable labels and their corresponding softmax scores:
```python
import torch
from transformers import AutoModelForSequenceClassification, AutoTokenizer
model = AutoModelForSequenceClassification.from_pretrained("poltextlab/xlm-roberta-large-dutch-cap-v3")
tokenizer = AutoTokenizer.from_pretrained("xlm-roberta-large")
sentence = "This is an example."
inputs = tokenizer(sentence,
return_tensors="pt",
max_length=512,
padding="do_not_pad",
truncation=True
)
logits = model(**inputs).logits
probs = torch.softmax(logits, dim=1).tolist()[0]
probs = {model.config.id2label[index]: round(probability, 2) for index, probability in enumerate(probs)}
top3_probs = dict(sorted(probs.items(), key=lambda item: item[1], reverse=True)[:3])
print(top3_probs)
```
## Model performance
The model was evaluated on a test set of 6398 examples.<br>
Model accuracy is **0.83**.
| label | precision | recall | f1-score | support |
|:-------------|------------:|---------:|-----------:|----------:|
| 0 | 0.81 | 0.77 | 0.79 | 471 |
| 1 | 0.7 | 0.72 | 0.71 | 148 |
| 2 | 0.88 | 0.8 | 0.84 | 242 |
| 3 | 0.76 | 0.87 | 0.81 | 78 |
| 4 | 0.76 | 0.78 | 0.77 | 374 |
| 5 | 0.9 | 0.92 | 0.91 | 248 |
| 6 | 0.86 | 0.75 | 0.8 | 155 |
| 7 | 0.79 | 0.86 | 0.82 | 95 |
| 8 | 0.86 | 0.82 | 0.84 | 217 |
| 9 | 0.88 | 0.9 | 0.89 | 244 |
| 10 | 0.85 | 0.87 | 0.86 | 763 |
| 11 | 0.73 | 0.75 | 0.74 | 319 |
| 12 | 0.79 | 0.83 | 0.81 | 121 |
| 13 | 0.75 | 0.77 | 0.76 | 378 |
| 14 | 0.82 | 0.83 | 0.83 | 123 |
| 15 | 0.7 | 0.75 | 0.72 | 106 |
| 16 | 0.39 | 0.58 | 0.47 | 19 |
| 17 | 0.93 | 0.92 | 0.93 | 1136 |
| 18 | 0.86 | 0.84 | 0.85 | 903 |
| 19 | 0.64 | 0.75 | 0.69 | 72 |
| 20 | 0.86 | 0.82 | 0.84 | 186 |
| macro avg | 0.79 | 0.8 | 0.79 | 6398 |
| weighted avg | 0.84 | 0.83 | 0.83 | 6398 |
### Fine-tuning procedure
This model was fine-tuned with the following key hyperparameters:
- **Number of Training Epochs**: 10
- **Batch Size**: 8
- **Learning Rate**: 5e-06
- **Early Stopping**: enabled with a patience of 2 epochs
## Inference platform
This model is used by the [CAP Babel Machine](https://babel.poltextlab.com), an open-source and free natural language processing tool, designed to simplify and speed up projects for comparative research.
## Cooperation
Model performance can be significantly improved by extending our training sets. We appreciate every submission of CAP-coded corpora (of any domain and language) at poltextlab{at}poltextlab{dot}com or by using the [CAP Babel Machine](https://babel.poltextlab.com).
## Reference
Sebők, M., Máté, Á., Ring, O., Kovács, V., & Lehoczki, R. (2024). Leveraging Open Large Language Models for Multilingual Policy Topic Classification: The Babel Machine Approach. Social Science Computer Review, 0(0). https://doi.org/10.1177/08944393241259434
## Debugging and issues
This architecture uses the `sentencepiece` tokenizer. In order to use the model before `transformers==4.27` you need to install it manually.
If you encounter a `RuntimeError` when loading the model using the `from_pretrained()` method, adding `ignore_mismatched_sizes=True` should solve the issue.