- Import the needed Python packages and define a few constants:
import numpy as npimport tensorflow as tffrom tensorflow.contrib.rnn import *import matplotlib.pyplot as pltnum_neurons = 100num_inputs = 1num_outputs = 1symbol = 'goog' # amznepochs = 500seq_len = 20 learning_rate = 0.001
numpy (http://www.numpy.org) is the most popular Python library for n-dimensional array operation, and Matplotlib (https://matplotlib.org) is the leading Python 2D plotting library. We'll use numpy to process the dataset and Matplotlib to visualize the stock prices and predictions. num_neurons is the number of neurons the RNN, or more accurately, an RNN cell, has at each time step - each neuron receives both the input element ...
