Update app.py

app.py

@@ -1,30 +1,40 @@
 import gradio as gr
-from pytube import extract
 import os
 import pandas as pd
-import string
+from pytube import extract
 import re
+import string
+import pickle
 import nltk
 import nltk.sentiment.util
 from nltk.corpus import stopwords
 from nltk.stem import WordNetLemmatizer 
-from textblob import TextBlob
-import numpy as np
-import math
 
-nltk.download('stopwords')
+from sklearn.metrics import accuracy_score
 
+from keras.preprocessing.text import Tokenizer
+from keras.preprocessing.sequence import pad_sequences
+from tensorflow import keras
+
+import xgboost
+
+nltk.download(stopwords)
 sw = stopwords.words('english')
 lemmatizer = WordNetLemmatizer()
 
-
 ## get YouTube ID
 def getID(url):
     print("Getting YouTube ID...")
     return extract.video_id(url)
 
-## function to clean comments
+## download comments
+def downloadComments(videoID):
+    print("Downloading Comments...")
+    os.system("youtube-comment-downloader --youtubeid=" + videoID + " --output Comments/" + videoID + ".json")
+
+# function to clean comments
 def clean_text(text):
+    
     # remove symbols and Emojis
     text = text.lower()
     text = re.sub('@', '', text)
@@ -53,20 +63,33 @@ def clean_text(text):
     
     return text
 
-## download comments
-def downloadComments(videoID):
-    print("Downloading Comments...")
-    os.system("youtube-comment-downloader --youtubeid=" + videoID + " --output Comments/" + videoID + ".json")
-
-
 def getSentenceTrain():
     # open sentences_train file
     sentences_train_f = open('../Deep learning/pickles/sentences_train.pickle', "rb")
     sentences_train = pickle.load(sentences_train_f)
     sentences_train_f.close()
     return sentences_train
-    
-## voting function
+
+# open pickle file
+# randFor_71_f = open('../Shallow machine learning/pickles/randFor_71.pickle', "rb")
+# randFor_train = pickle.load(randFor_71_f)
+# randFor_71_f.close()
+
+SGD_74_f = open('../Shallow machine learning/pickles/SGD_74.pickle', "rb")
+SGD_train = pickle.load(SGD_74_f)
+SGD_74_f.close()
+
+# XGB_74_f = open('../Shallow machine learning/pickles/XGB_74.pickle', "rb")
+# XGB_train = pickle.load(XGB_74_f)
+# XGB_74_f.close()
+
+logreg_79_f = open('../Shallow machine learning/pickles/logreg_79.pickle', "rb")
+logreg_train = pickle.load(logreg_79_f)
+logreg_79_f.close()
+
+# get saved CNN model
+model = keras.models.load_model("../Deep learning/CNN_82")
+
 def vote(test_point, _test):
     print("Voting on video effectivess...\n")
     pos_weighting = []
@@ -75,11 +98,11 @@ def vote(test_point, _test):
     algos_score = 0
     
     algorithms = [
-             {'name': 'Random Forest', 'accuracy': accuracy_score(y_test, y_pred_RandFor)*100, 'trained': randFor_train},
-             {'name': 'SGD', 'accuracy': accuracy_score(y_test, y_pred_SGD)*100, 'trained': SGD_train},
-             {'name': 'XGBoost', 'accuracy':  accuracy_score(y_test, y_pred_SGD)*100, 'trained': XGB_train},
-             {'name': 'Logistic Regression', 'accuracy': accuracy_score(y_test, y_pred_logreg)*100, 'trained': logreg_train},
-             {'name': 'CNN', 'accuracy': CNN_accuracy*100, 'trained': model}
+            #  {'name': 'Random Forest', 'accuracy': 0.71*100, 'trained': randFor_train},
+             {'name': 'SGD', 'accuracy': 0.74*100, 'trained': SGD_train},
+            #  {'name': 'XGBoost', 'accuracy':  0.74*100, 'trained': XGB_train},
+             {'name': 'Logistic Regression', 'accuracy': 0.79*100, 'trained': logreg_train},
+             {'name': 'CNN', 'accuracy': 0.82*100, 'trained': model}
         ]
     
     for algo in algorithms:
@@ -104,7 +127,7 @@ def vote(test_point, _test):
         result = 'effective'
         confidence = pos_result
         
-    print("\nThis video is \033[1m" + result + "\033[0m with a confidence of \033[1m" + str(round(confidence*100,2)) + "% \033[0m")
+    return result
 
 def quantizeEffectiveness(url):
     # 1. Get YouTube ID
@@ -128,7 +151,7 @@ def quantizeEffectiveness(url):
     # 4. Create test dataframe
     test = pd.DataFrame([[videoID]], columns=['VideoID'])
     
-    # 5. Get documnets (pre-processd comments)
+    # 5. Get documents (pre-processd comments)
     test_documents = []
     comment = pd.read_csv("Processed Comments/" + videoID + "_all_words.csv")
     test_documents.append(list(comment["0"]))
@@ -151,14 +174,14 @@ def quantizeEffectiveness(url):
     _test = pad_sequences(tokenizer.texts_to_sequences(test_sentence), padding='post', maxlen=100)
     
     # 10. Vote on video effectiveness
-    vote(test_point,_test)
+    vote_result = vote(test_point,_test)
     
-def greet(name):
-    # vote = quantizeEffectiveness("https://www.youtube.com/watch?v=DhhVr5iLF-c")
-    if not os.exists('comments'):
-        os.mkdir('comments')
+def greet(url):
+    vote_result = quantizeEffectiveness(url)
+    if not os.exists('Comments'):
+        os.mkdir('Comments')
     
-    return "Hello " + name + "!!"
+    return vote_result
 
 iface = gr.Interface(fn=greet, inputs="text", outputs="text")
 iface.launch()