Fix noncustom model issue

api/DataProcesser.py

@@ -52,14 +52,18 @@ def handle_classify(self, df, classifier):
  #return classifier_switcher.get(classifier, lambda: "Invalid Classifier")(df)
 
  def get_pipeline(self, model_name):
+ if os.path.exists('assets/tweet_emotions.csv'):
+ prefix = ''
+ else:
+ prefix = 'public/'
  if model_name=="emotion_pipeline.pkl":
- df = pd.read_csv('assets/tweet_emotions.csv')
+ df = pd.read_csv(prefix + 'assets/tweet_emotions.csv')
  train_data, test_data, train_target, test_target = train_test_split(df['content'], df['sentiment'], test_size=0.2, shuffle=True)
  elif model_name=="hate_speech.pkl":
- df = pd.read_csv('assets/nb_hatespeech.csv', sep=';')
+ df = pd.read_csv(prefix + 'assets/nb_hatespeech.csv', sep=';')
  train_data, test_data, train_target, test_target = train_test_split(df['comment'], df['isHate'], test_size=0.2, shuffle=True)
  elif model_name=="text_classification_pipeline.pkl":
- df = pd.read_csv('assets/nb_news.csv')
+ df = pd.read_csv(prefix + 'assets/nb_news.csv')
  train_data, test_data, train_target, test_target = train_test_split(df['short_description'], df['category'], test_size=0.2, shuffle=True)
  else:
  with open(f'api/models/{model_name}', 'rb') as file:
@@ -109,9 +113,12 @@ def pretrained_predict(self, df, pipeline, model_name = None):
  texts_to_predict = [str(text) for text in texts_to_predict]
 
  predictions = pipeline.predict(texts_to_predict)
- label_predictions = label_encoder.inverse_transform(predictions)
 
- df['output_column'] = label_predictions
+ if model_name:
+ label_predictions = label_encoder.inverse_transform(predictions)
+ df['output_column'] = label_predictions
+ else:
+ df['output_column'] = predictions
 
  return df