The general idea of the project is to encapsulate general Machine Learning Model Training, it's training simulation and finally to display the inference with general Exploratory Data Analysis libraries.
The user will have the ability to explore data mapping, model insights, and tracking the ML Lifecycle with a matter of simple clicks!
This progressive web app integrates the Machine Learning life cycle along with the development of model inference, which can have the ability to handle data re-collection, data selection, performing one-step EDA - which includes both statistical and visual analysis and getting results for better features selection.
Moreover we have incorporated real-time tracking of analytical model training performance[Only for Deep Learning Models].
Classes are easily imported without any hassles.
They include:
- Load_Data
- PreProcessing
- Evaluations
- Statistical Analysis
- Real-Time Tracking of ML models and log retrival
- Custom Inference
Streamlit is used to include all the models in our web application to productionalize it.
Further, we have implemented the real-time tracking with the help of Google Firebase. The Firebase Realtime Database is a cloud-hosted NoSQL database that lets you store and sync data between your users in realtime. Hence the data in fed from firebase in-real-time and it is visualized on the graphs.
We come to the end of the project, where we have created a 'One Click EDA' The EDA consists of :
univariate graphical,
multivariate graphical,
univariate non-graphical,
and multivariate non-graphical
but we have consolidated all the above into a single functionality.
To use the repo with main app dashboard and run inferences, please follow the guidelines below:
-
Clone the Repository:
$ git clone https://github.com/harshgeek4coder/Workspace_Labs.git -
Enter the directory:
$ cd Workspace_Labs -
Install the requirements:
$ pip install -r requirements.txt -
For running on CLI - For the main Streamlit Dash Board, use the inference file as follows:
$ streamlit run app.py
With this, we hope you liked the project, if you did please make sure to leave a star, that will surely boost our morale! Thanks!
NOTE : This project is made with the intention that an ordinary person can make sense out of, in other words, one who does not have any prior Machine Learning knowledge.