balamurugan1603
commited on
Commit
•
fc2a686
1
Parent(s):
997f997
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Named Entity Recognition using Transformers
|
2 |
+
This is a Fine-tuned version of BERT using HuggingFace transformers to perform Named Entity Recognition on Text data. BERT is a state-of-the-art model with attention mechanism as underlying architecture trained with masked-language-modeling and next-sentence-prediction objectives, used for various tasks including Question answering systems, Text Summarization, etc... which can also perform token classification tasks such as NER with great performance.
|
3 |
+
|
4 |
+
# Dataset
|
5 |
+
**CoNLL-2003** :
|
6 |
+
The shared task of CoNLL-2003 concerns language-independent named entity recognition. We will concentrate on four types of named entities: persons, locations, organizations, and names of miscellaneous entities that do not belong to the previous three groups.<br><br>
|
7 |
+
**Link** : https://huggingface.co/datasets/conll2003
|
8 |
+
|
9 |
+
# Using this fine-tuned version
|
10 |
+
|
11 |
+
From python, download the whole pipeline and use it instantly using the following code :
|
12 |
+
```
|
13 |
+
from transformers import pipeline
|
14 |
+
|
15 |
+
# Loading the pipeline from hub
|
16 |
+
# Pipeline handles the preprocessing and post processing steps
|
17 |
+
model_checkpoint = "balamurugan1603/bert-finetuned-ner"
|
18 |
+
namedEntityRecogniser = pipeline(
|
19 |
+
"token-classification", model=model_checkpoint, aggregation_strategy="simple"
|
20 |
+
)
|
21 |
+
```
|
22 |
+
|
23 |
+
Reference for using this pipeline to find NER tags can be found in this <a href="https://github.com/balamurugan1603/Named-Entity-Recognition-using-Tranformers/blob/main/named-entity-recognition-using-transfer-learning.ipynb">notebook</a>.
|