Final Project for Embedded Machine Learning of Heidelberg University. Group5 task is replicate the ADMM pruning and implement it specific LeNet-5 and AlexNet.
Management Document: https://www.notion.so/fffrozen/Embedded-Machine-Learning-Final-Project-16034c82d99a41a0b3cefdaef3a72d8c?pvs=4
Dicionary Structure:
src/
: Contains the source code for the projectADMM_Toolbox/
: Contains the ADMM toolboxSimplePruning_Toolbox/
(Optional) : Contains the simple pruning toolbox such as 1. Enegy Efficiency-Aware Pruning, 2. Structured-Preserved Pruning. These methods are not major task of the project, but it is good comparison with ADMM.LeNet5/
: Contains the LeNet5 implementationAlexNet/
: Contains the AlexNet implementation
data/
: Contains the data for the projectresults/
: Contains the results of the projectdocs/
: Contains the documentation of the projectreferences/
: Contains the references of the projectdocker/
: Contains the docker files for the project
TODOs: Implement AlexNet for ImageNet Implement LeNet5 for MNIST Implement ADMM Pruning for LeNet5 Implement ADMM Pruning for AlexNet