from tensorflow.keras.models import load_model, Sequential, Model from models import Models from sklearn.metrics import confusion_matrix, f1_score, precision_score, recall_score from text_preprocessing import TextPreprocessing import json import os from csv_reader import CSVReader from datetime import datetime from tensorflow.keras import backend as K from sklearn.metrics import classification_report textPreProcessing = TextPreprocessing() if __name__ == "__main__": embeddings = np.load('text_embedding.npy', allow_pickle=True) texts = [] sentiments = [] good = CSVReader.dataframe_from_txt("WordNetAffectEmotionLists/joy.txt") good = good.texts texts = texts = np.append(texts, np.array(good)) sentiments = np.append(sentiments, [1] * len(good)) surprise = CSVReader.dataframe_from_txt( "WordNetAffectEmotionLists/surprise.txt") surprise = surprise.texts texts = np.append(texts, np.array(surprise)) sentiments = np.append(sentiments, [2] * len(surprise)) sad = CSVReader.dataframe_from_txt("WordNetAffectEmotionLists/sadness.txt") sad = sad.texts texts = np.append(texts, np.array(sad)) sentiments = np.append(sentiments, [-1] * len(sad)) fear = CSVReader.dataframe_from_txt("WordNetAffectEmotionLists/fear.txt") fear = fear.texts texts = np.append(texts, np.array(fear))