Update app.py

app.py

@@ -14,6 +14,10 @@ negative_preds = grouped_predictions.get_group(0)
 
 predictions_df['Prediction'] = predictions_df['Prediction'].map({0: 'Negative', 1: 'Neutral', 2: 'Positive'})
 
+# Load training data set
+dataset = load_dataset("eengel7/sentiment_analysis_training", split='train')
+training_df = pd.DataFrame(dataset)
+random_sample = {}
 
 def article_selection(sentiment):
     if sentiment == "Positive":
@@ -38,29 +42,32 @@ def article_selection(sentiment):
 
 def manual_label():
     # Selecting random row from batch data
+    global random_sample
     random_sample = predictions_df.sample()
-    print('hey')
-    #random_sample_ds = Dataset.from_pandas(random_sample)
-    #random_sample.to_csv('/Users/torileatherman/Github/ID2223_scalable_machine_learning/news_articles_sentiment/sample.csv', index=False)
-    #random_sample_ds.push_to_hub('torileatherman/sample', index=False)
     random_headline = random_sample['Headline_string'].iloc[0]
     random_prediction = random_sample['Prediction'].iloc[0]
+    
     return random_headline, random_prediction
 
 
 def thanks(sentiment):
-    labeled_sentiments = []
-    labeled_sentiments.append(sentiment)
-    #counter = len(labeled_sentiments)
-    #counter = str(counter)
-    #login(token = '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')
-    #create_repo("torileatherman/"+counter+"labeled_data")
-    labeled_sentiments = pd.DataFrame(labeled_sentiments, columns = ['Manual Predictions'])
-    labeled_sentiments.to_csv('/Users/torileatherman/Github/ID2223_scalable_machine_learning/news_articles_sentiment/manual_labels.csv', index=False)
-    #labeled_sentiments = Dataset.from_pandas(labeled_sentiments)
-    #labeled_sentiments.push_to_hub("torileatherman/"+counter+"labeled_data")
-    return f"""Thank you for making our model better!"""
+
+    # Create int label 
+    mapping = gender = {'Negative': 0,'Neutral': 1, 'Positive':2}
+    sentiment = int(mapping[sentiment])
+    
+    global training_df
+    # Append training data set
+    training_df = training_df.append({'Sentiment': sentiment, 'Headline_string': random_sample['Headline_string'].iloc[0], 'Headline': random_sample['Headline'].iloc[0] }, ignore_index=True)
     
+    # Upload training data set
+    ds = Dataset.from_pandas(training_df)
+    try: 
+        ds.push_to_hub("eengel7/sentiment_analysis_training")
+    except StopIteration:
+        pass
+
+    return f"""Thank you for making our model better! """
 
 description1 =  '''
         This application recommends news articles depending on the sentiment of the headline.