|
--- |
|
datasets: |
|
- DFKI-SLT/few-nerd |
|
language: |
|
- en |
|
metrics: |
|
- f1=0.87 |
|
- precision |
|
- recall |
|
--- |
|
# HuggingsaurusRex/bert-base-uncased-for-mountain-ner |
|
|
|
## Purpose |
|
Detect mountain names in text using token classification. |
|
|
|
## Usage |
|
```python |
|
from transformers import AutoModelForTokenClassification, AutoTokenizer, pipeline |
|
|
|
# Load model and tokenizer |
|
model = AutoModelForTokenClassification.from_pretrained('huggingsaurusRex/bert-base-uncased-for-mountain-ner') |
|
tokenizer = AutoTokenizer.from_pretrained('huggingsaurusRex/bert-base-uncased-for-mountain-ner') |
|
|
|
# Create NER pipeline |
|
ner = pipeline('ner', model=model, tokenizer=tokenizer) |
|
|
|
# Perform inference |
|
res = ner("I spent days climbing the Mount Everest.") |
|
print(res) |
|
``` |
|
|
|
## Architecture |
|
The model is a BERT-based token classification model fine-tuned on the Few-NERD dataset. |
|
|
|
## Results |
|
- F1-Score: 0.87 |
|
- Precision: 0.84 |
|
- Recall: 0.91 |
|
|
|
## Direct Link |
|
[HuggingsaurusRex/bert-base-uncased-for-mountain-ner](https://huggingface.co/huggingsaurusRex/bert-base-uncased-for-mountain-ner) |
|
|
|
|
