import streamlit as st 
import streamlit.components.v1 as stc 
import pandas as pd 
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics.pairwise import cosine_similarity,linear_kernel
import re
import pandas as pd 
import numpy as np 
import matplotlib.pyplot as plt 
import neattext.functions as nfx
from sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer
from sklearn.metrics.pairwise import cosine_similarity
import warnings
st.set_page_config(layout="wide", initial_sidebar_state="expanded")

warnings.filterwarnings("ignore")
data= pd.read_csv("udemy_courses.csv")
data["titre_OK"]= data.course_title.apply(nfx.remove_stopwords)
data["titre_OK"]= data.course_title.apply(nfx.remove_special_characters)
countVec= CountVectorizer()
cv= countVec.fit_transform(data.titre_OK)
df=data
matrice_cosine= cosine_similarity(cv)

def recommend_course2(title, numrec=10):
    try:
        pattern = re.compile(re.escape(title), re.IGNORECASE)
        matching_courses = df['course_title'].apply(lambda x: bool(pattern.search(x)))    
        index = df[matching_courses].index[0]
        scores = list(enumerate(matrice_cosine[index]))
        sorted_scores = sorted(scores, key=lambda x: x[1], reverse=True)
        selected_course_index = [i[0] for i in sorted_scores[1:]]
        selected_course_score = [i[1] for i in sorted_scores[1:]]
        rec_df = df.iloc[selected_course_index]
        rec_df['Similarity_Score'] = selected_course_score

        final_recommended_courses = rec_df[["course_title","level", "subject","Similarity_Score"]]
    except:
        final_recommended_courses= pd.DataFrame({"data": "Aucune Recommendaion disponible!"},index=[0])
        
    return final_recommended_courses.head(numrec)



def main():

	st.title("Système de Recommandation de Cours")

	menu = ["Accueil","Recommendations","A propos"]
	choice = st.sidebar.selectbox("Menu",menu)


	if choice == "Accueil":
		st.subheader("Accueil")
		st.dataframe(df.head(10))


	elif choice == "Recommendations":
		st.subheader("Recommendations de formation")
		search_term = st.text_input("cours")
		num_of_rec = st.sidebar.number_input("Nombre de cours",4,30,7)
		if st.button("Recommendations"):
			if search_term:
				st.write(recommend_course2(search_term, num_of_rec))
	else:
		st.subheader("A propos")
		st.text("Keyce @2024")


if __name__ == '__main__':
	main()