This release has been a month in the making. The official release notes are here. A Thinking in JAX tutorial has been introduced in this release for newbies to get a better hang of JAX based numerical computing. The main sparse recovery algorithm introduced in this release is SPGL1 (Spectral Projected Gradient L1). Testing the algorithm under different conditions was an interesting experience. It finally gave me a strong motivation to port Sparco problems into CR-Sparse....
Python has excellent support for digital signal processing of ECG signals. In this post, we shall explore some basic capabilities for plotting ECG data and doing some basic signal processing for identifying the R peaks inside the signals. We shall use NumPy for basic numerical computations and matplotlib for plotting. # numerical computations import numpy as np # plotting from matplotlib import pyplot as plt SciPy includes a sample electrocardiogram signal....
This article is an early draft. Docker is a technology to created isolated environments on a host computer. One can compare Docker to virtual machines on one hand and language specific virtual environments like conda on the other hand. Virtual machines provide complete hardware virtualization. Tools like conda can provide isolated environments for individual applications but they are specific to a particular language (like Python). Docker is language independent. It provides software isolation....
How to get started? Modern web application development is a huge field. There are hundreds of technologies in use to build a high quality application. This post is a roadmap to get your grooves going to become a productive web developer. Frontend and Backend Any web application has two parts. One part is the piece of code that is running inside the browser of your user. Another part is the piece of code that is running inside your own servers....
Paper on CR-Sparse published in JOSS My paper titled “CR-Sparse: Hardware accelerated functional algorithms for sparse signal processing in Python using JAX” has been published in “The Journal of Open Source Software”. Citation: Kumar, S., (2021). CR-Sparse: Hardware accelerated functional algorithms for sparse signal processing in Python using JAX. Journal of Open Source Software, 6(68), 3917, https://doi.org/10.21105/joss.03917