Spaces:
Sleeping
Sleeping
Add app.py
Browse files
app.py
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import numpy as np
|
3 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
4 |
+
|
5 |
+
# Load the tokenizer and model
|
6 |
+
model_name = "NLPTeamIITGN/finetuned_llama_sentiment_sst2"
|
7 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
8 |
+
model = AutoModelForSequenceClassification.from_pretrained(model_name)
|
9 |
+
|
10 |
+
# Define a label map
|
11 |
+
label_map = {0: "Negative", 1: "Positive"}
|
12 |
+
|
13 |
+
# Function to predict sentiment
|
14 |
+
def predict_sentiment(review):
|
15 |
+
if not review.strip():
|
16 |
+
return "No input provided", 0.0
|
17 |
+
inputs = tokenizer(review, return_tensors="pt", truncation=True, padding=True, max_length=512)
|
18 |
+
outputs = model(**inputs)
|
19 |
+
logits = outputs.logits.detach().numpy()
|
20 |
+
predicted_label = np.argmax(logits, axis=-1)[0]
|
21 |
+
predicted_class = label_map[predicted_label]
|
22 |
+
probability = np.max(logits, axis=-1)[0]
|
23 |
+
return predicted_class, round(float(probability), 2)
|
24 |
+
|
25 |
+
# Gradio Interface
|
26 |
+
interface = gr.Interface(
|
27 |
+
fn=predict_sentiment,
|
28 |
+
inputs=gr.Textbox(lines=3, placeholder="Enter a movie review here...", label="Movie Review"),
|
29 |
+
outputs=[
|
30 |
+
gr.Label(label="Sentiment"),
|
31 |
+
gr.Number(label="Confidence Score")
|
32 |
+
],
|
33 |
+
title="Sentiment Analysis for Movie Reviews",
|
34 |
+
description="Enter a movie review, and the model will classify it as Positive or Negative with a confidence score."
|
35 |
+
)
|
36 |
+
|
37 |
+
# Launch the Gradio app
|
38 |
+
interface.launch()
|