src : has the wrappers required to use the models or functionalities

models : holds the trained models

pipeline_ui : hold the GUI for data_pipeline

figures : output directory

data : input directory and may hold raw/processed/interim data

requirements.txt : holds the packages to be installed to execute the pipeline

run.py : program to call the UI

**** Create miniconda environment(to enable single project based environment that doesn't mix up with other project environments) ****

Install miniconda

cd <drive_name>:

cd python

md <project_folder_name>

cd <project_folder_name>

conda create --prefix ./env pandas numpy matplotlib scikit-learn

conda activate <dive_name>:\Python<project_folder_name>\env

conda install jupyter

jupyter notebook

import pandas as pd import numpy as np import matplotlib.pyplot as plt import sklearn

ctrl+c

conda deactivate

**** To install requirements after pulling updated repo use->> pip isntall -r requirements.txt ****

**** Use google colaboratory for training for large datasets that requires lot of GPU ****