To Do.txt

## ML project

# To do:

Initialize github repository 

1. Build a website (Check Streamlit)
	- design a webpage (pages, specs,...)
	- flask, html, css integration (The look)
	- run it on a localhost

2. Integrate backend for ML model predictions
	- handle inputs/outputs
	- check for graphs and visualization

3. Data collection and preparation

	* ETL (Extract, transform and load)
			- Extraction (python)
			- Transform (python)
			- Load 
	* EDA (Explanatory Data Analysis)/Descriptive analysis

4. Build a model
	- build model for prediciton
    	* Encoding categorical variables
    	* PCA analysis and dimension reduction
    	* Try different models/algorithms
	- build model for recomendation
	- evaluate models (R2 score, Mae and Mse)
	- save models

5. Combine models and webpage
	- combine functions such as train/test/predict
	- create sort of dashboard for prediction (https://www.chartjs.org/)

6. Dockerize the application

7. Deploy the application on Amazon or Streamlit?