Project Proposal Geant4-FastSim - ML pipeline optimization using Kubeflow

This is the tentative proposal submission for the Geant4-FastSim - Building an ML pipeline for fast shower simulation. This proposal discusses a detailed approach to implementing the Kubeflow End to End Pipeline for the existing framework. The proposal consists of a synopsis discussing how the proposed approach was thought and what were the key insights that motivated the steps mentioned. It further discusses the task that will be encountered throughout the Project timeline and the Major Deliverables that would mark the completion of the project. The Major tasks have been visited in detail in the proposal and these are used to determine the objectives for the weekly progress timeline in the project. The proposal tries to cover up all the adopted methodology's significant detail and come up with the most ideal way to tackle each and every task. Deliverables: End to End Deployed ML Pipeline on Kubeflow Meta Logger component throughout the pipeline to store and analyze the results Experimenting with existing and new algorithmic approaches that might improve the performance, throughput of the modeling pipeline Creation of a connector component for the transition from training phase into the inference phase. Iterative development of the Pipeline into a generalized workflow that is capable of scaling to newer versions of data as well support the integration of newer algorithmic experimentation Well documented results and analysis for motivating additional focus on the pipeline as well motivating the use of this pipeline for different projects \