README.md

---
license: apache-2.0
language:
- en
- fr
- es
- multilingual
widget:
- text: "Critical levels of out of  school children were reported, with 72% of respondents pointing to moderate to high numbers of primary school age not accessing <mask>"
---

# HumBert 

HumBert (Humanitarian Bert) is a [XLM-Roberta](https://huggingface.co/xlm-roberta-base) model trained on humanitarian texts - approximately 50 million textual examples (roughly 2 billion tokens) from public humanitarian reports, law cases and news articles.
Data were collected from three main sources: [Reliefweb](https://reliefweb.int/), [UNHCR Refworld](https://www.refworld.org/) and [Europe Media Monitor News Brief](https://emm.newsbrief.eu/).
Although XLM-Roberta was trained on 100 different languages, this fine-tuning was performed on three languages, English, French and Spanish, due to the impossibility of finding a good amount of such kind of  humanitarian data in other languages.


## Intended uses

To the best of our knowledge, HumBert is the first language model adapted on humanitarian topics, which often use a very specific language, making adaptation to downstream tasks (such as dister responses text classification) more effective.
This model is primarily aimed at being fine-tuned on tasks such as sequence classification or token classification.

## Benchmarks

Soon...

## Usage

Here is how to use this model to get the features of a given text in PyTorch:

```python
from transformers import AutoTokenizer, AutoModelForMaskedLM
tokenizer = AutoTokenizer.from_pretrained('nlp-thedeep/humbert')
model = AutoModelForMaskedLM.from_pretrained("nlp-thedeep/humbert")
# prepare input
text = "YOUR TEXT"
encoded_input = tokenizer(text, return_tensors='pt')
# forward pass
output = model(**encoded_input)
```