We will be using a different approach from the Keras model used earlier in the project to build this model. To build a more complex model, we will use Tensorflow to write each layer from scratch. Tensorflow gives us as Data Scientists and Deep Learning Engineers more fine-tuned control over our model architecture. 

For this model, we will use the code in this block to create 2 placeholders which will store the input and output values:

import tensorflow as tf
import pickle
from tensorflow.contrib import rnn

    def build(self, input_number, sequence_length, layers_number, units_number, output_number):
        self.x = tf.placeholder("float", [None, sequence_length, input_number])
        self.y = tf.placeholder("float", [None, output_number])
        self.sequence_length = sequence_length

Next, we need to store the weights and bias in variables we create:

        self.weights = {
            'out': tf.Variable(tf.random_normal([units_number, output_number]))
        }
        self.biases = {
            'out': tf.Variable(tf.random_normal([output_number]))
        }

        x = tf.transpose(self.x, [1, 0, 2])
        x = tf.reshape(x, [-1, input_number])
        x = tf.split(x, sequence_length, 0)

We build this model using multiple LSTM layers with the basic LSTM cells assigning each layer with the specified number of cells as shown in the figure below:

        lstm_layers = []
        for i in range(0, layers_number):
            lstm_layer = rnn.BasicLSTMCell(units_number)
            lstm_layers.append(lstm_layer)

        deep_lstm = rnn.MultiRNNCell(lstm_layers)

        self.outputs, states = rnn.static_rnn(deep_lstm, x, dtype=tf.float32)

        print "Build model with input_number: {}, sequence_length: {}, layers_number: {}, " 
              "units_number: {}, output_number: {}".format(input_number, sequence_length, layers_number,
                                                           units_number, output_number)
# This method is using to dump the model configurations 
        self.save(input_number, sequence_length, layers_number, units_number, output_number)