---
metrics:
- precision
- recall
- f1
pipeline_tag: token-classification
tags:
- material science
---

# SpringerMaterials BERT NER multilingual

This model was fine-tuned on NER-labeled user queries from [Springer Materials](https://materials.springer.com/). The entity types supported by this model are: `SUBSTANCE`, `PROPERTY`.
The fine-tuning data is available [here](https://doi.org/10.6084/m9.figshare.24592428).

## Input examples

* 🇬🇧 `ethylene's phase diagram`
* 🇩🇪 `Phasendiagramm des Ethylens`
* 🇷🇺 `фазовая диаграмма этилена`

## How to use this model

```python
from transformers import pipeline

model_checkpoint = "perevalov/SMatBERT-NER-multilingual"
token_classifier = pipeline(
    "ner", model=model_checkpoint, aggregation_strategy="simple"
)

token_classifier("boiling point of water")
```