import pandas as pd
import tensorflow as tf
#import numpy as np
from tensorflow import keras
from tensorflow.keras import layers
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler


df = pd.read_csv('aat.us.txt', names=["Data","Open","High","Low","Close","Volume"], usecols=[0,1,2,3,4,5] , parse_dates=True, index_col=0)


t_train, val_train, t_target, val_target = train_test_split(df, df['Close'] , test_size=0.2)



#normalização
std = StandardScaler()

t_train = std.fit_transform(t_train)
t_target = std.fit_transform(t_target)
val_train = std.fit_transform(val_train)
val_target = std.fit_transform(val_target)

#train model

def build_model():
  model = keras.Sequential([
   layers.Dense(64, activation='relu', input_shape=[len(t_train.keys())]),
   layers.Dense(64, activation='relu'),
   layers.Dense(1)
  ])

  optimizer = tf.keras.optimizers.RMSprop(0.001)

  model.compile(loss='mse',
                optimizer=optimizer,
                metrics=['mae', 'mse'])
  return model


model = build_model()

model.summary()