Update 'create-dataset.py'

create-dataset.py

@@ -1,6 +1,10 @@
 import pandas
 import os
+
+from keras.applications.densenet import layers
+
 from sklearn.model_selection import train_test_split
+import tensorflow
 
 
 CUTOFF = int(os.environ['CUTOFF'])
@@ -10,13 +14,16 @@ video_games = pandas.read_csv('./ium_z434686/Video_Games_Sales_as_at_22_Dec_2016
                               engine='python',
                               encoding='ISO-8859-1',
                               sep=',')
+
 # DROP NA FIELDS
 video_games = video_games.dropna()
+video_games = video_games.drop(video_games.columns[[0, 1, 3, 4, 13, 14, 15]], axis=1)
 
 # CUT OFF DATASET TO X LINES
 video_games = video_games.sample(CUTOFF)
 
-X, Y = video_games, video_games
+X = video_games.copy()
+Y = pandas.DataFrame(video_games.pop('User_Score'))
 
 # SPLIT BETWEEN DEV, TRAINS, AND TEST
 X_train, X_temp, Y_train, Y_temp = train_test_split(X, Y, test_size=0.3, random_state=1)
@@ -25,3 +32,31 @@ X_dev, X_test, Y_dev, Y_test = train_test_split(X_temp, Y_temp, test_size=0.3, r
 X_train.to_csv('X_train.csv', index=False)
 X_dev.to_csv('X_dev.csv', index=False)
 X_test.to_csv('X_test.csv', index=False)
+Y_test.to_csv('Y_test.csv', index=False)
+Y_train.to_csv('Y_train.csv', index=False)
+Y_dev.to_csv('Y_dev.csv', index=False)
+
+train_data_x = pandas.read_csv('./X_train.csv')
+
+
+train_data_x = pandas.read_csv('./X_train.csv')
+
+
+games_all = train_data_x.copy()
+games_predict = train_data_x.pop('User_Score')
+normalize = layers.Normalization()
+normalize.adapt(games_all)
+
+norm_games_model = tensorflow.keras.Sequential([
+  normalize,
+  layers.Dense(64),
+  layers.Dense(1)
+])
+
+norm_games_model.compile(
+    loss=tensorflow.keras.losses.MeanSquaredError(),
+    optimizer=tensorflow.keras.optimizers.Adam())
+
+norm_games_model.fit(games_all, games_predict, epochs=500)
+
+norm_games_model.save('test')