Upload 9 files

.gitattributes

@@ -33,3 +33,5 @@ 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
+maindataInstagram.csv filter=lfs diff=lfs merge=lfs -text
+maindataYoutube.csv filter=lfs diff=lfs merge=lfs -text

age_range.csv

@@ -0,0 +1,20 @@
+Audience_age_range,Encoded_Value
+10-15,0
+15-20,1
+20-25,2
+25-30,3
+30-35,4
+35-40,5
+40-45,6
+45-50,7
+5-10,8
+50-55,9
+55-60,10
+60-65,11
+65-70,12
+70-75,13
+75-80,14
+80-85,15
+85-90,16
+90-95,17
+95-100,18

algo_instagram.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e11ce9eb65ebb6f0a364c5557a672c8374ea6882e5d620d8321227dfdf09f14b
+size 305964536

algo_youtube.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c42fdee89021b447a64256cb0887894a65e4dada9818545e916952140462df48
+size 194283087

app.py

@@ -0,0 +1,113 @@
+import streamlit as st
+import numpy as np
+import pandas as pd
+import pickle
+
+youtube_model = pickle.load(open('algo_youtube.pkl', 'rb'))
+instagram_model = pickle.load(open('algo_instagram.pkl', 'rb'))
+maindataYoutube =  pd.read_csv("maindataYoutube.csv")
+maindataInstagram =  pd.read_csv("maindataInstagram.csv")
+
+age_df = pd.read_csv("age_range.csv")
+category_df =  pd.read_csv("category.csv")
+country_df =  pd.read_csv("country.csv")
+
+st.title("Recommendation")
+
+def format_gender(option):
+       
+    if option == 0:
+        return "Male"
+    elif option == 1:
+        return "Female"
+    else:
+        return option
+    
+def format_age(option):
+    age = age_df.loc[age_df['Encoded_Value'] == option, 'Audience_age_range'].values[0]
+    return age
+    
+    
+def format_category(option):
+    content_category = category_df.loc[category_df['Encoded_Value'] == option, 'Content_category'].values[0]
+    return content_category
+    
+    
+def format_country(option):
+    country = country_df.loc[country_df['Encoded_Value'] == option, 'Audience_country'].values[0]
+    return country
+    
+
+target_gender = st.selectbox(label= "Select Target Gender", options= [0,1], format_func= format_gender)
+platform = st.selectbox(label= "Select Platform", options= ["Instagram", "Youtube"])
+category = st.selectbox(label= "Select Category", options= category_df['Encoded_Value'], format_func= format_category) # edit
+budget = st.slider(label="Select Budget", min_value= 10000, max_value= 800000) #edit
+min_engagement_rate = st.slider(label="Select Minimum Engagement Range", min_value= 1, max_value= 5) #edit
+target_country = st.selectbox(label="Select Country" , options= country_df['Encoded_Value'],format_func= format_country ) #edit
+target_age_range = st.selectbox(label="Select Target Age Range" ,options= age_df['Encoded_Value'], format_func= format_age) #edit
+
+def filtered_influencer_youtube():
+    filtered_youtubers = []
+    for _,youtuber in maindataYoutube.iterrows():
+        if youtuber['Engagement_rate'] >= min_engagement_rate \
+                and youtuber['Majority_audience_gender'] == target_gender \
+                and youtuber['Audience_country'] == target_country \
+                and youtuber['Audience_age_range'] == target_age_range \
+                and youtuber['Cost_per_post'] <= budget:
+            filtered_youtubers.append((youtuber["YouTube_channel_name"],youtuber['Content_category'], youtuber['Cost_per_post']))
+
+    return filtered_youtubers
+
+def filtered_influencer_instagram():
+    filtered_insta = []
+    for _, insta in maindataInstagram.iterrows():
+        if insta['Engagement_rate'] >= min_engagement_rate \
+                and insta['Majority_audience_gender'] == target_gender \
+                and insta['Audience_country'] == target_country \
+                and insta['Audience_age_range'] == target_age_range \
+                and insta['Cost_per_post'] <= budget:
+            filtered_insta.append((insta["Influencer_insta_username"],insta['Content_category'], insta['Cost_per_post']))
+           
+    return filtered_insta
+
+
+def get_recommendations_with_age_range_filter_youtube():
+    
+    filtered = filtered_influencer_youtube()
+    dataset1 = [tup[:3] for tup in filtered]
+    predictions = {}
+    for uid, iid, category in dataset1:
+        pred = youtube_model.predict(uid, iid)
+        predictions[uid] = pred.est
+    sorted_predictions = sorted(predictions.items(), key=lambda x: x[1], reverse=True)
+    return sorted_predictions[:10]
+    
+def get_recommendations_with_age_range_filter_instagram():
+    filtered = filtered_influencer_instagram()
+    dataset1 = [tup[:3] for tup in filtered]
+    predictions = {}
+    for uid, iid, category in dataset1:
+        pred = instagram_model.predict(uid, iid)
+        predictions[uid] = pred.est
+    sorted_predictions = sorted(predictions.items(), key=lambda x: x[1], reverse=True)
+    return sorted_predictions[:10]
+    
+
+def checkPlatform():
+    
+    if platform == "Youtube":
+        youtubers = get_recommendations_with_age_range_filter_youtube()
+        for youtuber, rating in  youtubers:
+             youtuber_name = maindataYoutube.loc[maindataYoutube['YouTube_channel_name'] == youtuber , 'channel_name'].iloc[0]
+             st.success(youtuber_name + ' :thumbsup:')       
+        
+        
+    else:
+        influencers = get_recommendations_with_age_range_filter_instagram()
+        for influencer, rating in influencers:
+            influencer_name = maindataInstagram.loc[maindataInstagram['Influencer_insta_username'] == influencer, 'Influencer_insta_id'].iloc[0]
+            st.success(influencer_name + ' :thumbsup:') 
+        
+    
+
+trigger = st.button('Recommend ', on_click=checkPlatform)    

category.csv

@@ -0,0 +1,29 @@
+Content_category,Encoded_Value
+Animals,0
+Art,1
+Beauty,2
+Business & Careers,3
+Cinema/Theatre,4
+Comedy,5
+Dance,6
+Education,7
+Family,8
+Fashion,9
+Fitness,10
+Food & Cooking,11
+Games,12
+Inspirational quotes,13
+Lifestyle,14
+Literature & Journalism,15
+Management & Marketing,16
+Media,17
+Medical,18
+Modeling,19
+Music,20
+Photography,21
+Real Estate,22
+Retail,23
+Spirituality,24
+Sports,25
+Technology,26
+Travel,27

country.csv

@@ -0,0 +1,52 @@
+Audience_country,Encoded_Value
+Algeria,0
+Argentina,1
+Australia,2
+Austria,3
+Bangladesh,4
+Belgium,5
+Bhutan,6
+Brazil,7
+Canada,8
+Chile,9
+China,10
+Colombia,11
+Costa Rica,12
+Denmark,13
+Egypt,14
+France,15
+Germany,16
+Greece,17
+Hong Kong,18
+India,19
+Indonesia,20
+Iraq,21
+Ireland,22
+Italy,23
+Japan,24
+Malaysia,25
+Mexico,26
+Morocco,27
+Netherlands,28
+New Zealand,29
+Nigeria,30
+Norway,31
+Pakistan,32
+Peru,33
+Philippines,34
+Poland,35
+Portugal,36
+Romania,37
+Russia,38
+Saudi Arabia,39
+Singapore,40
+South Africa,41
+South Korea,42
+Spain,43
+Sweden,44
+Switzerland,45
+Thailand,46
+United Arab Emirates,47
+United Kingdom,48
+United States,49
+Vietnam,50

maindataInstagram.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5af3ede92a72880b638c1440db9ecc1e591dfe8401d51d4298aabf581fa94cbe
+size 74914142

maindataYoutube.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c24eecf4b4ec7227a56a728675f076795bd7de88ce746b13ef61f104bb357a8
+size 40408901

requirements.txt

@@ -0,0 +1,5 @@
+streamlit==1.20.0
+numpy>=1.9.2
+scikit-learn>=0.18
+pandas 
+surprise