---
language:
- fr
datasets:
- amazon_reviews_multi
license: mit
---

# flaubert_small_cased_sentiment

This is a `flaubert_small_cased` model finetuned for sentiment analysis on product reviews in French. It predicts the sentiment of the review, from `very_negative` (1 star) to `very_positive` (5 stars).

This model is intended for direct use as a sentiment analysis model for French product reviews, or for further finetuning on related sentiment analysis tasks.

## Training data

The training data consists of the French portion of `amazon_reviews_multi`, supplemented with another 140,000 similar reviews.

## Accuracy

The finetuned model was evaluated on the French test set of `amazon_reviews_multi`.

- Accuracy (exact) is the exact match on the number of stars.
- Accuracy (off-by-1) is the percentage of reviews where the number of stars the model predicts differs by a maximum of 1 from the number given by the human reviewer. 


| Language | Accuracy (exact) | Accuracy (off-by-1) |
| -------- | ---------------------- | ------------------- |
| French  | 61.56%                 | 95.66%

## Contact 

[NLP Town](https://www.nlp.town) offers a suite of sentiment models for a wide range of languages, including an improved multilingual model through [RapidAPI](https://rapidapi.com/nlp-town-nlp-town-default/api/multilingual-sentiment-analysis2/). 

Feel free to contact us for questions, feedback and/or requests for similar models.