Hands-On Industrial Internet of Things

Defining the model

We know nothing about NASA engines, nor how they operate. We only have data. We have a lot of it, however, so we will use a data-driven model. The model will be based on an RNN using Keras and TensorFlow. The basic steps to build our model are:

1. Building the RNN and scaling the data set in the interval [0,1]
2. Building the data set to train the model
3. Training the model
4. Testing the model

Below the code of the algorithm:

1. On the first 5 lines Keras and the Scikit-Learn packages jave been imported. The function prepare_dataset() scales the data, and the function build_model() builds the model:

# fit the modelimport mathfrom keras.models import Sequentialfrom keras.layers import Densefrom sklearn.preprocessing import MinMaxScaler ...