Home

Welcome to the MLP-OpenCL wiki!

In this Wiki I am going to go through some of the theoretical knowledge needed for working on this project.

Mostly, my focus is going to be to get to know a little bit about OpenCL so there is entire Wiki chapter dedicated to it.

1. Multi-Layer Perceptron

1.1 Theoretical basis

1.2 Implementation overview

2. Multiprocessor systems

2.1 Types of multiprocessor systems

2.2 How to parallelize MLP

2.3 Different libraries and methods for parallelization

2.3.1 OpenCL

3. Methodology

Here, the some sort of "diary" will be kept.

Refferences that i used while working on the project:

1. Simple Tensorflow code for training the MLP for MNIST dataset in Tensorflow https://www.digitalocean.com/community/tutorials/how-to-build-a-neural-network-to-recognize-handwritten-digits-with-tensorflow

2. Nice tool for creating test images http://dot2pic.com/

3. OpenCL error codes https://gist.github.com/bmount/4a7144ce801e5569a0b6

4. Youtube videos on OpenCL that helped a lot https://www.youtube.com/watch?v=8D6yhpiQVVI&list=PLhqBhHU0mKh9zeei8cdnEe4I5e6ZMNlqy

5. Nvidia guide https://www.cs.cmu.edu/afs/cs/academic/class/15668-s11/www/cuda-doc/OpenCL_Best_Practices_Guide.pdf

6. Makefile tutorial/guide https://opensource.com/article/18/8/what-how-makefile