In this post we put together all the building blocks covered in previous posts to create a convolution neural network, using numpy, and test it on the MNIST hand-written digits classification task.
This post covers the derivations of back-propagation in a convolution layer, with numpy implementations.
This post covers the implementation of pooling layers in a convolutional neural network using numpy.
This post will share some knowledge of 2D and 3D convolutions in a convolution neural network (CNN), and 3 implementations all done using pure `numpy` and `scipy`.
This post will share some basic knowledge of an artificial neural network and how to create one from scratch using only numpy. We will build a classification network to classify hand-written digits from the MNIST dataset.
This post shares a Python implementation to compute the Silhouette Values in a clustering analysis.
The Ramer-Douglas-Peucker (RDP) algorithm is a curve simplification method. To apply it on coordinates defined by latitudes/longitudes, we need to replace the Cartesian geometry with a spherical one.
The convolution functions in `scipy` do not work well with missing data. We create a 2D convolution function that allows a controllable tolerance to missing values. It is first implemented in Fortran, then using `scipy` in an FFT approach.
Peak prominence can be used to identify relatively organized regional maxima while filtering out local disturbances.