polished main pipeline

textclassifier/TextClassifier.py

@@ -109,7 +109,7 @@ class TextClassifier:
         """
         Analyzes the sentiment of a text using OpenAI.
         :param text: string of the tweet text.
-        :param sentiment:
+        :param sentiment: string of the sentiment.
         :return:
         """
         # assert 1 == 2, "Måste fixa prompt innan denna metod körs"
@@ -275,28 +275,37 @@ class TextClassifier:
         """
         Classifies the topics/sentiments of a user's tweets.
         #We presume that all tweets inside the twitterdata.csv file are already classified.
-        :return: Pandas dataframe
+        :return: None
         """
-        # temp_df = pd.DataFrame(
-        #     columns=['id', 'tweet', 'date', 'user_id', 'username', 'urls', 'nlikes', 'nreplies', 'nretweets'])
+        # Check if file exists, if not, create it
         if os.path.exists(filename):
             already_classified_df = pd.read_csv(filename)
+            print("Already classified tweets: {}".format(already_classified_df.shape[0]))
             # Create a temporary df where values from already_classified_df that are not it self.df are stored
             temp_df = self.df[self.df['id'].isin(already_classified_df['id'])]
             # Remove rows from self.df that are not in already_classified_df
             self.df = self.df[~self.df['id'].isin(already_classified_df['id'])]
+            print("Classifying topic of {} tweets...".format(self.df.shape[0]))
+            self.df = self.classify_topics_of_tweets()
+            print("Classifying sentiment of {} tweets...".format(self.df.shape[0]))
+            self.df = self.classify_sentiment_of_tweets()
+            print("Writing to csv...")
+            self.df_to_csv(filename)
+            # Concatenate temp_df and self.df
+            self.df = pd.concat([temp_df, self.df], ignore_index=True)
+            print("Appended {}.".format(filename))
+            return None
         else:
             print("No csv file found. Continuing without removing already classified tweets.")
-
-        self.df = self.classify_topics_of_tweets()
-        self.df = self.classify_sentiment_of_tweets()
-        # self.df = self.analyze_sentiment_of_tweets()
-
-        self.df_to_csv(filename)
-        # Concatenate temp_df and self.df
-        self.df = pd.concat([temp_df, self.df], ignore_index=True)
-        print("Done.")
-        return 0
+            print("Classifying topics...")
+            self.df = self.classify_topics_of_tweets()
+            print("Classifying sentiments...")
+            self.df = self.classify_sentiment_of_tweets()
+            # self.df = self.analyze_sentiment_of_tweets()
+            print("Writing to csv file...")
+            self.df_to_csv(filename)
+            print("Created {}.".format(filename))
+            return None
 
 
     def __repr__(self):