Sruthi Karicheri, Suparna Bhattacharjee(https://github.com/suparnabh), Jennie Ran (https://github.com/jran14)

Scope of the Project::point_down:

1. To build a prediction model to determine the downhole pressure of a test well.
2. To understand the interdependencies of the different variables in the dataset

Task Breakdown::point_down:

a. Extraction and Understanding of the production and subsurface dataset.

Analyze the various production parameters from each well and establish correlations and variances.

b. Transforming the dataset :notebook_with_decorative_cover: NOTEBOOK - jupyternotebook

– Performing some data wrangling and clean up
- exploratory data analysis
    - Pair plots
    - Heat map
    - Histogram

c. Design training and testing parameters/wells.

📔 NOTEBOOK - jupyternotebook

- linear regression 
- Basic Linear model with Lasso shrinkage
- random forest regression model  

📔 NOTEBOOK - Final Notebooks_Hyppertuning and Plots/grid_randomized_searchcv.ipynb

    -Gridsearch CV
    -Tune Lasso Model
    -Tune Randomforest Model 

📔 NOTEBOOK - NeuralAnalysis.ipynb

    -Neural Network Model
    -Hyperas tuning of NN

📔 NOTEBOOK individual well analysis

    - EDA_individualwells.ipynb (for future analysis)

Final Report :- Project 3 Report.docx

Presentation slide :- ml_presentation.pptx

Our Website Machine Learning of the Equinor Volve Field Production Data