An exploration into an efficient FIR filter implementation with single-instruction, multiple data in C++. Inspired by Jan Wilczek's work.
License: GNU Lesser General Public License v2.1
C++ 70.04%CMake 4.19%C 7.29%Makefile 18.45%TypeScript 0.03%
fir-simd's Introduction
Hi, I'm Desh Iyer.
I'm a B.Tech. CSE Hons. student with a passion for the profound and transformative Free Energy Principle (FEP). This concept acts as a connective thread across the diverse domains of cognitive science, mathematics, philosophy, and AI, resonating with my belief that understanding the interplay between internal models and the external world is key to advancing artificial intelligence among other fields of study.
During my internship at KEK, Japan, I witnessed the power of distributed computing in handling massive datasets, which undoubtedly broadened my perspective and fueled my passion for exploring the intersection of FEP and computational physics. My work at KEK involves the development of the BelleDIRAC framework, which, employed in high-energy physics experiments, relies on distributed computing to efficiently store an unfathomable amount of data across servers in institutions all over the world.
This correlation between the FEP and distributed computing is more than just a research endeavor for me. It's a fascinating journey into the heart of AI, where the integration of insights from cognitive science, mathematics, and philosophy holds the promise of creating truly autonomous and sentient machines. This, to me, is not just about shaping the future of AI; it's about being a part of a breakthrough that transforms the way we perceive and interact with intelligent systems. I'm excited about contributing to this convergence of disciplines to hopefully make a meaningful impact on the landscape of artificial intelligence.
This README is a list of all the repositories of code pertaining to a multitude of domains. These repos are either on my GitHub or on other sites.
This repository contains analysis of the Telecom Churn Dataset, exploring customer behavior patterns and predicting churn using machine learning techniques
A repository that documents how a model is trained in one notebook and its weights exported and loaded into another notebook with Pytorch. The example is a BERT model for document classification
A repository that takes a pre-trained T5 model from HuggingFace and fine-tunes it on a dataset of labelled newspaper articles for document classification
A repository that documents the right way to create virtual environments and link them to Jupyter kernels for deep learning projectsA repository to document learning Next.js, one of the most popular front-end frameworks in JavaScript to build user-interfaces