|
--- |
|
license: mit |
|
language: |
|
- fr |
|
library_name: transformers |
|
inference: false |
|
pipeline_tag: feature-extraction |
|
--- |
|
# CamemBERT-L10 |
|
|
|
This model is a pruned version of the pre-trained [CamemBERT](https://huggingface.co/camembert-base) checkpoint, obtained by [dropping the top-layers](https://doi.org/10.48550/arXiv.2004.03844) from the original model. |
|
|
|
 |
|
|
|
## Usage |
|
|
|
You can use the raw model for masked language modeling (MLM), but it's mostly intended to be fine-tuned on a downstream task, especially one that uses the whole sentence to make decisions such as text classification, extractive question answering, or semantic search. For tasks such as text generation, you should look at autoregressive models like [BelGPT-2](https://huggingface.co/antoinelouis/belgpt2). |
|
|
|
You can use this model directly with a pipeline for [masked language modeling](https://huggingface.co/tasks/fill-mask): |
|
|
|
```python |
|
from transformers import pipeline |
|
|
|
unmasker = pipeline('fill-mask', model='antoinelouis/camembert-L10') |
|
unmasker("Bonjour, je suis un [MASK] modèle.") |
|
``` |
|
|
|
You can also use this model to [extract the features](https://huggingface.co/tasks/feature-extraction) of a given text: |
|
|
|
```python |
|
from transformers import AutoTokenizer, AutoModel |
|
|
|
tokenizer = AutoTokenizer.from_pretrained('antoinelouis/camembert-L10') |
|
model = AutoModel.from_pretrained('antoinelouis/camembert-L10') |
|
|
|
text = "Remplacez-moi par le texte de votre choix." |
|
encoded_input = tokenizer(text, return_tensors='pt') |
|
output = model(**encoded_input) |
|
``` |
|
|
|
## Variations |
|
|
|
CamemBERT has originally been released in base (110M) and large (335M) variations. The following checkpoints prune the base variation by dropping the top 2, 4, 6, 8, and 10 pretrained encoding layers, respectively. |
|
|
|
| Model | #Params | Size | Pruning | |
|
|--------------------------------------------------------------------|:-------:|:-----:|:-------:| |
|
| [CamemBERT-base](https://huggingface.co/camembert-base) | 110.6M | 445MB | - | |
|
| | | | | |
|
| **CamemBERT-L10** | 96.4M | 386MB | -13% | |
|
| [CamemBERT-L8](https://huggingface.co/antoinelouis/camembert-L8) | 82.3M | 329MB | -26% | |
|
| [CamemBERT-L6](https://huggingface.co/antoinelouis/camembert-L6) | 68.1M | 272MB | -38% | |
|
| [CamemBERT-L4](https://huggingface.co/antoinelouis/camembert-L4) | 53.9M | 216MB | -51% | |
|
| [CamemBERT-L2](https://huggingface.co/antoinelouis/camembert-L2) | 39.7M | 159MB | -64% | |
|
|
|
## Citation |
|
|
|
For attribution in academic contexts, please cite this work as: |
|
|
|
```bibtex |
|
@online{louis2023, |
|
author = 'Antoine Louis', |
|
title = 'CamemBERT-L10: A Pruned Version of CamemBERT', |
|
publisher = 'Hugging Face', |
|
month = 'october', |
|
year = '2023', |
|
url = 'https://huggingface.co/antoinelouis/camembert-L10', |
|
} |
|
``` |
