Track large dataset with Git LFS

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+self_help_books.csv filter=lfs diff=lfs merge=lfs -text

app.py

@@ -1,7 +1,113 @@
 import gradio as gr
+import pandas as pd
+import numpy as np
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
 
-def greet(name):
-    return "Hello " + name + "!!"
+def recommend_books(concern, top_n=5, reviews_per_book=2):
+    # Load and preprocess data
+    df = pd.read_csv('self_help_books.csv')
+    
+    # Create TF-IDF vectors from reviews
+    tfidf = TfidfVectorizer(stop_words='english')
+    review_vectors = tfidf.fit_transform(df['Review'].fillna(''))
+    concern_vector = tfidf.transform([concern])
+    
+    # Calculate similarity scores
+    similarities = cosine_similarity(concern_vector, review_vectors).flatten()
+    
+    # Get top books based on review similarity
+    top_indices = np.argsort(similarities)[-top_n:][::-1]
+    recommended_books = df.iloc[top_indices].copy()
+    
+    # Add helpful and harmful reviews
+    for idx, row in recommended_books.iterrows():
+        book_reviews = df[df['Book'] == row['Book']]
+        
+        # Get helpful reviews
+        helpful_reviews = book_reviews.nlargest(reviews_per_book, 'Helpful_Ratio')['Review'].tolist()
+        recommended_books.at[idx, 'Helpful Reviews'] = helpful_reviews
+        
+        # Get critical reviews
+        harmful_reviews = book_reviews.nsmallest(reviews_per_book, 'Helpful_Ratio')['Review'].tolist()
+        recommended_books.at[idx, 'Harmful Reviews'] = harmful_reviews
+        
+    return recommended_books
 
-demo = gr.Interface(fn=greet, inputs="text", outputs="text")
-demo.launch()
+def recommend_authors(concern, top_n=5):
+    df = pd.read_csv('self_help_books.csv')
+    
+    # Calculate author metrics
+    author_stats = df.groupby('author_clean').agg({
+        'Helpful_Ratio': ['mean', 'count']
+    }).reset_index()
+    
+    author_stats.columns = ['author_clean', 'helpful_ratio', 'review_count']
+    
+    # Filter authors with minimum reviews
+    min_reviews = 5
+    author_stats = author_stats[author_stats['review_count'] >= min_reviews]
+    
+    # Get top and bottom authors
+    good_authors = author_stats.nlargest(top_n, 'helpful_ratio')
+    risky_authors = author_stats.nsmallest(top_n, 'helpful_ratio')
+    
+    return good_authors, risky_authors
+
+def recommend_for_concern(concern, num_books=5, num_reviews=2):
+    """Wrapper function to format recommendations for Gradio"""
+    books_df = recommend_books(concern, top_n=num_books, reviews_per_book=num_reviews)
+    good_authors, risky_authors = recommend_authors(concern, top_n=num_books)
+    
+    # Format book recommendations
+    book_output = "=== RECOMMENDED BOOKS ===\n\n"
+    for _, book in books_df.iterrows():
+        book_output += f"📚 {book['Book']}\n"
+        book_output += f"👤 Author: {book['Author']}\n"
+        book_output += f"⭐ Rating: {book['Star_Rating']}\n"
+        book_output += f"💰 Price: ${book['Price']}\n"
+        book_output += f"📊 Helpful Ratio: {book['Helpful_Ratio']:.2f}\n"
+        
+        if book['Helpful Reviews']:
+            book_output += "\n✅ Helpful Reviews:\n"
+            for review in book['Helpful Reviews']:
+                book_output += f"• {review}\n"
+                
+        if book['Harmful Reviews']:
+            book_output += "\n⚠️ Critical Reviews:\n"
+            for review in book['Harmful Reviews']:
+                book_output += f"• {review}\n"
+        
+        book_output += "\n" + "-"*50 + "\n\n"
+
+    # Format author recommendations
+    author_output = "=== RECOMMENDED AUTHORS ===\n\n"
+    author_output += "✅ Authors Likely to be Helpful:\n"
+    for _, author in good_authors.iterrows():
+        author_output += f"• {author['author_clean']} (Helpful ratio: {author['helpful_ratio']:.2f})\n"
+    
+    author_output += "\n⚠️ Authors to Approach with Caution:\n"
+    for _, author in risky_authors.iterrows():
+        author_output += f"• {author['author_clean']} (Helpful ratio: {author['helpful_ratio']:.2f})\n"
+
+    return book_output + "\n\n" + author_output
+
+# Create the Gradio interface
+iface = gr.Interface(
+    fn=recommend_for_concern,
+    inputs=[
+        gr.Textbox(label="What concern or fear would you like help with?", placeholder="e.g. I'm a lonely teenager"),
+        gr.Slider(minimum=1, maximum=10, value=5, step=1, label="Number of recommendations"),
+        gr.Slider(minimum=1, maximum=5, value=2, step=1, label="Reviews per book")
+    ],
+    outputs=gr.Textbox(label="Recommendations", lines=20),
+    title="Self-Help Book Recommender",
+    description="Get personalized book recommendations based on your concerns or fears.",
+    examples=[
+        ["I'm a lonely teenager", 5, 2],
+        ["I'm worried about my career", 5, 2],
+        ["I have anxiety about the future", 5, 2]
+    ]
+)
+
+iface.launch()

self_help_books.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8acadc7a62c6f88ce0cde40fdbf41b1b4ee20d98dd7e519b2228b6dfb1d6e5b
+size 62107596