diff --git a/predictions.py b/predictions.py
index 8c8e433..d3ddf95 100644
--- a/predictions.py
+++ b/predictions.py
@@ -1,12 +1,12 @@
-import tensorflow
+import keras
+import numpy as np
+import tensorflow as tf
 import pandas as pd
 
-model = tensorflow.keras.models.load_model('model.h5')
-X_test_data = pd.read_csv("X_test.csv").astype(float)
-Y_test_data = pd.read_csv("Y_test.csv").astype(float)
+test_data = pd.read_csv("adult_test.csv")
 
-model.evaluate(X_test_data, Y_test_data)
+model = keras.models.load_model("model.h5")
 
-predictions = model.predict(X_test_data)
+predictions = model.predict(test_data)
 
-predictions.to_csv('predictions.csv', index=False)
+np.savetxt("predictions.csv", predictions, delimiter=",")
diff --git a/script.py b/script.py
index 46f4da0..e7fdad8 100644
--- a/script.py
+++ b/script.py
@@ -53,9 +53,9 @@ def check_if_data_set_has_division_into_subsets(file_name):
 
 
 def get_statistics(data):
-	train_data = pd.read_csv("X_train.csv", dtype={"income": "category"})
-	dev_data = pd.read_csv("X_dev.csv", dtype={"income": "category"})
-	test_data = pd.read_csv("X_test.csv", dtype={"income": "category"})
+	train_data = pd.read_csv("adult_train.csv", dtype={"income": "category"})
+	dev_data = pd.read_csv("adult_dev.csv", dtype={"income": "category"})
+	test_data = pd.read_csv("adult_test.csv", dtype={"income": "category"})
 
 	print("Wielkość zbioru: ", len(data))
 	print("Wielkość zbioru treningowego: ", len(train_data))
@@ -106,34 +106,31 @@ def clean(data):
 
 
 def train_dev_test(data):
-	X = data.copy()
-	y = pandas.DataFrame(data.pop('education-num'))
-	X_train, X_temp, Y_train, Y_temp = train_test_split(X, y, test_size=0.3, random_state=1)
-	X_dev, X_test, Y_dev, Y_test = train_test_split(X_temp, Y_temp, test_size=0.3, random_state=1)
+	train_data, test_data = train_test_split(data, test_size=0.3, random_state=42)
 
-	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)
-	return X_train, X_dev, X_test
+	test_data, dev_data = train_test_split(test_data, test_size=0.33, random_state=42)
+
+	train_data.to_csv("adult_train.csv", index=False)
+	dev_data.to_csv("adult_dev.csv", index=False)
+	test_data.to_csv("adult_test.csv", index=False)
+
+	return train_data, dev_data, test_data
 
 
 def create_model():
-	data = pd.read_csv('X_train.csv')
+	data = pd.read_csv('adult_train.csv')
 	X = data.copy()
 	y = data["education-num"]
 	X_train_encoded = pd.get_dummies(X)
 	y_train_cat = to_categorical(y)
 	model = Sequential()
 	model.add(Dense(64, activation='relu', input_dim=X_train_encoded.shape[1]))
-	model.add(Dense(17, activation='softmax'))
+	model.add(Dense(17, activation='sigmoid'))
 	model.compile(optimizer='adam',
-				  loss='categorical_crossentropy',
+				  loss='binary_crossentropy',
 				  metrics=['accuracy'])
 	model.fit(X_train_encoded, y_train_cat, epochs=10, batch_size=32, validation_data=(X_train_encoded, y_train_cat))
-	model.save('model.h5')
+	model.save('model.joblib')
 
 
 if __name__ == '__main__':