da.py

# -*- coding: utf-8 -*-
"""Da.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/1Kg9-C_Fif3yO8FuXT84Tci-3ItuLsS7p
"""

#Library install
!pip install transformers sentence-transformers gradio

import pandas as pd
from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_similarity
import gradio as gr

# Load the dataset
df = pd.read_csv('/content/courses.csv')  # Replace with actual path to courses.csv

# Load a pre-trained sentence transformer model
model = SentenceTransformer('all-MiniLM-L6-v2')

# Create a combined column for embedding (e.g., title + description + keywords)
df['combined_text'] = df['title'] + " " + df['description'] + " " + df['keywords']
course_embeddings = model.encode(df['combined_text'].tolist(), convert_to_tensor=True)

def search_courses(user_query):
    # Encode the user query
    query_embedding = model.encode(user_query, convert_to_tensor=True)

    # Compute cosine similarities between the query and each course embedding
    similarities = cosine_similarity(
        query_embedding.cpu().detach().numpy().reshape(1, -1),
        course_embeddings.cpu().detach().numpy()
    )

    # Get indices of top matching courses (top 5 results)
    top_matches = similarities.argsort()[0][-5:][::-1]

    # Retrieve top matching courses
    results = [{"title": df.iloc[i]["title"], "description": df.iloc[i]["description"]} for i in top_matches]
    return results

# Define Gradio function for user interaction
def gradio_search(query):
    results = search_courses(query)
    return results

# Set up Gradio interface
iface = gr.Interface(
    fn=gradio_search,
    inputs="text",
    outputs="json",
    title="Smart Course Search",
    description="Find the most relevant courses based on your query."
)

# Launch the app (for local testing or deploying in Hugging Face Spaces)
iface.launch()