sentiment-analysis-oid / app /sentiment_analysis.py
santanche's picture
feat (app): initial setup
47f370d
raw
history blame contribute delete
613 Bytes
from transformers import pipeline
class SentimentAnalysis:
def __init__(self):
# Initialize the sentiment analysis pipeline
self.pipe = pipeline("text-classification", model="tabularisai/multilingual-sentiment-analysis")
def classify(self, text: str):
# Use the pipeline to classify the input text
result_set = self.pipe(text)
result = result_set[0]
analysis = result["label"].lower()
result["value"] = "positive" if "positive" in analysis \
else "negative" if "negative" in analysis \
else "neutral"
return result