Upload 3 files

.gitattributes

@@ -32,3 +32,4 @@ 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
+justin_rf_model.sav filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,67 @@
+import streamlit as st
+import pickle
+from datetime import datetime
+from sentence_transformers import SentenceTransformer
+import numpy as np
+import re
+
+loaded_scaler = pickle.load(open('minMaxScaler.sav', 'rb'))
+loaded_model = pickle.load(open('justin_rf_model.sav', 'rb'))
+vectorizer = SentenceTransformer('all-MiniLM-L6-v2')
+
+def featurize(tweet, tweet_date, tweet_time):
+    # featurize date
+    #date = year + '-' + month + '-' + day
+    #date = datetime.strptime(date, '%Y-%m-%d')
+    year = float(tweet_date.year)
+    month = float(tweet_date.month)
+    day = float(tweet_date.day)
+    hr = float(tweet_time.hour)
+    minutes = float(tweet_time.minute)
+    weekDay = float(tweet_date.weekday())
+    
+    # preprocess tweets
+    tweet = re.sub(r'http\S+', 'url', tweet)
+    
+    # count the number of accounts tagged and hashtags mentioned in tweet
+    tagCount = float(len(re.findall(r"@(\w+)", tweet)))
+    hashTagsCount = float(len(re.findall(r"#(\w+)", tweet)))
+    
+    # vectorize data
+    x1 = np.array([tagCount, hashTagsCount, minutes, hr, day, weekDay, month, year ])
+    x1 = loaded_scaler.transform(x1.reshape(1,-1))
+    x2 = vectorizer.encode(tweet)
+    inp_vec = np.concatenate([x2, x1.flatten()]).reshape(1,-1)
+    
+    return inp_vec
+
+def getOutput(inp_vec):
+    output = loaded_model.predict(inp_vec)
+    return output[0]
+
+def main():
+
+	st.title("Welcome to tweet engagement predictor")
+	with st.form("my_form"):
+		tweet = st.text_input('Enter a tweet')
+		tweet_date = st.date_input("Enter the date of tweeting", 
+			value = datetime(2018,1,1,0,0),
+			min_value=datetime(2015,1,1,0,0),
+			max_value=datetime(2021,12,12,23,59))
+		tweet_time = st.time_input('Enter the time of tweeting')
+
+		# Every form must have a submit button.
+		submitted = st.form_submit_button("Submit")
+		
+	if submitted: 
+		inp_vec = featurize(tweet, tweet_date, tweet_time)
+		output = getOutput(inp_vec)
+		if output == 1 : 
+			st.write('Given tweet will get low engagment - less than 8800 ')
+		elif output == 2 : 
+			st.write('Given tweet will get moderate engagment - retweets in the range of 8800 to 24000')
+		elif output == 3 : 
+			st.write('Given tweet will get high engagment - more than 24000 retweets ')
+	      
+if __name__ == '__main__' : 
+	main()

justin_rf_model.sav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11704df4905726dbaf20ac6c7aa4bcec32046bdbf0b6b1ee332541aba2c55005
+size 17903631