Update app.py

app.py

@@ -1,5 +1,6 @@
 import streamlit as st
 from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
+from PIL import Image, ImageDraw, ImageFont
 
 # Load the tokenizer and model
 tokenizer = AutoTokenizer.from_pretrained("OatNapat/finetuned_yelp")
@@ -8,6 +9,28 @@ model = AutoModelForSequenceClassification.from_pretrained("OatNapat/finetuned_y
 # Create a sentiment analysis pipeline with the explicit tokenizer
 nlp = pipeline("sentiment-analysis", model=model, tokenizer=tokenizer)
 
+# Function to convert text to an image
+def text_to_image(text):
+    # Create a blank image with a white background
+    image = Image.new("RGB", (500, 100), "white")
+    draw = ImageDraw.Draw(image)
+
+    # Define the font and font size
+    font = ImageFont.load_default()
+    font_size = 20
+
+    # Calculate text width and height
+    text_width, text_height = draw.textsize(text, font)
+
+    # Calculate the position to center the text in the image
+    x = (image.width - text_width) / 2
+    y = (image.height - text_height) / 2
+
+    # Draw the text on the image
+    draw.text((x, y), text, fill="black", font=font)
+
+    return image
+
 st.title("Sentiment Analysis App")
 user_input = st.text_input("ป้อนประโยคเพื่อวิเคราะห์ความรู้สึก:")
 if user_input:
@@ -30,3 +53,6 @@ if user_input:
     st.write(f"Sentiment: {sentiment_explanation}")
     st.write(f"Confidence: {sentiment_score:.4f}")
 
+    # Convert the sentiment explanation to an image
+    sentiment_image = text_to_image(sentiment_explanation)
+    st.image(sentiment_image, caption="Sentiment Explanation", use_column_width=True)