Create app.py

# Step 1: Import the necessary toolkits
# We need 'gradio' to build the web app and 'transformers' to use the Hugging Face model.
import gradio as gr
from transformers import pipeline

# Step 2: Load our AI Model
# We create a 'pipeline' which is a simple way to use a pre-trained model.
# We tell it the task ("sentiment-analysis") and the specific model we chose.
# The first time the app runs, it will download the model. This might take a minute.
print("Loading model...")
sentiment_pipeline = pipeline("sentiment-analysis", model="distilbert-base-uncased-finetuned-sst-2-english")
print("Model loaded!")

# Step 3: Define the function for the app
# This function will take some text as input and use the model to predict the sentiment.
def analyze_sentiment(text):
result = sentiment_pipeline(text)
# The model returns a label ('POSITIVE' or 'NEGATIVE') and a confidence score.
# We'll return the whole result so we can see both.
return result

# Step 4: Create the Gradio Web App Interface
# This is where we design the look and feel of our app.
app = gr.Interface(
fn=analyze_sentiment,
inputs=gr.Textbox(placeholder="Enter a sentence here..."),
outputs="json", # We'll use a JSON output to see the label and score clearly
title="Sentiment Analyzer",
description="Type in a sentence to see if its sentiment is POSITIVE or NEGATIVE. This app uses a DistilBERT model from Hugging Face.",
examples=[
["The new Star Wars movie was incredible!"],
["I am not happy with the customer service."],
["The weather today is just okay."]
]
)

# Step 5: Launch the app!
app.launch()

app.py

@@ -0,0 +1,38 @@
+# Step 1: Import the necessary toolkits
+# We need 'gradio' to build the web app and 'transformers' to use the Hugging Face model.
+import gradio as gr
+from transformers import pipeline
+
+# Step 2: Load our AI Model
+# We create a 'pipeline' which is a simple way to use a pre-trained model.
+# We tell it the task ("sentiment-analysis") and the specific model we chose.
+# The first time the app runs, it will download the model. This might take a minute.
+print("Loading model...")
+sentiment_pipeline = pipeline("sentiment-analysis", model="distilbert-base-uncased-finetuned-sst-2-english")
+print("Model loaded!")
+
+# Step 3: Define the function for the app
+# This function will take some text as input and use the model to predict the sentiment.
+def analyze_sentiment(text):
+    result = sentiment_pipeline(text)
+    # The model returns a label ('POSITIVE' or 'NEGATIVE') and a confidence score.
+    # We'll return the whole result so we can see both.
+    return result
+
+# Step 4: Create the Gradio Web App Interface
+# This is where we design the look and feel of our app.
+app = gr.Interface(
+    fn=analyze_sentiment,
+    inputs=gr.Textbox(placeholder="Enter a sentence here..."),
+    outputs="json",  # We'll use a JSON output to see the label and score clearly
+    title="Sentiment Analyzer",
+    description="Type in a sentence to see if its sentiment is POSITIVE or NEGATIVE. This app uses a DistilBERT model from Hugging Face.",
+    examples=[
+        ["The new Star Wars movie was incredible!"],
+        ["I am not happy with the customer service."],
+        ["The weather today is just okay."]
+    ]
+)
+
+# Step 5: Launch the app!
+app.launch()