Python-Based TensorFlow

TensorFlow is an open-source software library for dataflow programming across a range of jobs. It is a representative math library, and it is also used for ML applications such as neural networks. It is used for both research and production at Google.

In this study, the Python-based TensorFlow development environment consists of Python 3.5.4, TensorFlow 1.8.0, Anaconda 4.2.0 64 bit, and Pycharm community edition 2017.2.4 on Windows 10 64 bit. In TensorFlow, the hypothesis function was

where X is an input matrix, W is a weight matrix, b is a bias matrix, and Y is an output matrix. Multinomial logistic regression, also known as softmax regression, was used as the activation function. We used one hot encoding that encoded categorical integer features. A cross-entropy cost function was used as the cost function, and a gradient descent optimizer was used to minimize the cost function iteratively. In order for gradient descent to work, we set the learning rate to 0.001 and the number of iterations to 300,000.