π DistilBert Drug NER by BlaikHole
π Overview
This repository provides a fine-tuned model trained on private drug NER data using quick still efficient DistilBert architecture. It can be used for Drug NER with 7 classes.
π¨ Model Outputs & Labels
The model identifies the following labels:
Label Name | Description |
---|---|
π₯ LABEL_0 > Dosage | Drug dosage/quantity mentioned in text. |
π© LABEL_1 > Drug name | Drug name including Opioids. |
π¦ LABEL_2 > Event | Event like overdose, death. |
π¨ LABEL_3 > Location | Location of possible event. |
πͺ LABEL_4 > Other (Outside) | Outside of rest 7 entities. |
π§ LABEL_5 > ROA | Route of Administration for drug usage like injection or oral use. |
β¬ LABEL_6 > Symptom | Medical condition. |
π« LABEL_7 > Temporal | Time, date or frequency. |
π Quick Usage
You can easily load and use this model with transformers
:
πΉ Named Entity Recognition (NER)
import torch
from transformers import pipeline, AutoModelForTokenClassification, AutoTokenizer
# Label Mapping
LABEL_MAP = {
0: "DOSAGE",
1: "DRUG_NAME",
2: "EVENT",
3: "LOCATION",
4: "OTHER",
5: "ROA",
6: "SYMPTOM",
7: "TEMPORAL",
}
# Model Name
MODEL_NAME = "blaikhole/distilbert-drug-ner"
# Load Model & Tokenizer
device = 0 if torch.cuda.is_available() else -1
model = AutoModelForTokenClassification.from_pretrained(MODEL_NAME).to("cuda" if device == 0 else "cpu")
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
# Load NER Pipeline
ner_pipeline = pipeline("ner", model=model, tokenizer=tokenizer, device=device)
def process_text(text):
entities = ner_pipeline(text)
results = [(entity["word"], LABEL_MAP.get(int(entity["entity"].split("_")[-1]), "OTHER")) for entity in entities]
return results
# Example Usage
if __name__ == "__main__":
sample_text = "The patient was prescribed oxycodone and experienced dizziness."
print(process_text(sample_text))
π¦ Installation
To use this model, install the required dependencies:
pip install transformers torch
π License
MIT
- Downloads last month
- 69
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for blaikhole/distilbert-drug-ner
Base model
distilbert/distilbert-base-uncased