Neural Network Tutorials - Herong's Tutorial Examples - v1.22, by Herong Yang
SciPy - Python Library for Mathematical Functions
This section provides a tutorial example on how to install Python 3 SciPy library on macOS computers. SciPy is widely used by Python users for mathematical functions required in neural network models.
What Is SciPy? SciPy is an open-source Python library for scientific computing developed initially by Travis Oliphant and now maintained by the SciPy community.
If you want to build neural network models in Python, you should install SciPy and get familiar with its functionalities by following this tutorial. This is because neural network models require lots of mathematical functions, which are provided in SciPy.
1. Install SciPy library using the "pip3" (Package Installer for Python 3) command:
herong$ python3 --version Python 3.8.0 herong$ sudo pip3 install scipy Collecting scipy Downloading https://files.pythonhosted.org/packages/... Installing collected packages: scipy Successfully installed scipy-1.5.0
3. Verify SciPy installation by importing "scipy" package, retrieving its version string, and calling some statistical functions.
herong$ python3 Python 3.8.0 (v3.8.0:fa919fdf25, Oct 14 2019, 10:23:27) >>> import scipy as sp >>> sp.v1.22 '1.5.0' >>> import numpy as np >>> s = np.random.rand(1000000) >>> from scipy import stats >>> stats.tmean(s) 0.5000266615822236 >>> stats.tvar(s) 0.08345433925168123
Note that:
Cool! You have SciPy library ready on your Python 3 environment for scientific computing.
For more readings on SciPy, visit SciPy documentation Website at https://docs.scipy.org/doc/scipy/reference/.
Table of Contents
Deep Playground for Classical Neural Networks
►Building Neural Networks with Python
NumPy - Python Library for Matrix operations
►SciPy - Python Library for Mathematical Functions
Simple Example of Neural Networks
TensorFlow - Machine Learning Platform
PyTorch - Machine Learning Platform
CNN (Convolutional Neural Network)
RNN (Recurrent Neural Network)
GAN (Generative Adversarial Network)