aashrith48
Engineer at heart with a keen interest in Artificial Intelligence and Object-Oriented Programming (OOP).
Texas A&M UniversityCollege Station
Pinned Repositories
Electrofacies-characterization-
This project involved characterizing the reservoir into different electrofacies based on well log data using Principal Component Analysis and Unsupervised Learning (K-Means clustering)
Neural-Net-using-sklearn-and-keras-with-shale-well-completions-data
Developed neural net using both keras library and sklearn library to predict the Estimated ultimate recovery (EUR) label from shale gas well completions data features
Dimensionality-reduction-of-completions-data
Using Principal cluster analysis (PCA) and Non-negative matrix(NMF) for dimensionality reduction on a completions dataset consisting of
Gas-Price
Visualizing Gas prices with matplotlib
Hierachial-clustering-and-Isolation-forest-on-Petroleum-Data
Used hierarchial clustering methods to perform unsupervised learning on GIP-EUR data to classify into cluster.
K-means-clustering-on-geological-dataset
Performing K-means clustering on features consisting of geological data, along with optimal cluster number
K-Nearest-neighbours-with-HR-dataset
Used K- nearest neighbours classifier with HR dataset as features to predict the chances of an employee quitting, and compared it with logistic regression
Linear-regression-on-Geologic-Data
Performed linear regression on geologic data with appropriate visualization for viewing correlations between features to drop features for prediction
Logistic-regression-on-HR-Data
Performed logistic regression on human resource data to for binary classification to predict whether an employee will quit a job or continue to stay.
LSTM-RNN-with-Frac-treatment-pressure-
Predicted frac treatment pressure with using Long short team memory type RNN (Recurrent neural network) where I used past 60 seconds of data to predict the pressure at the next second
aashrith48's Repositories
aashrith48/LSTM-RNN-with-Frac-treatment-pressure-
Predicted frac treatment pressure with using Long short team memory type RNN (Recurrent neural network) where I used past 60 seconds of data to predict the pressure at the next second
aashrith48/Neural-Net-using-sklearn-and-keras-with-shale-well-completions-data
Developed neural net using both keras library and sklearn library to predict the Estimated ultimate recovery (EUR) label from shale gas well completions data features
aashrith48/Random-forest-and-Decision-tree-regressors-with-geological-data
Used randomforest and decision tree regressor to predict Total Organic Content (TOC) from features consisting of geological data and compared the performance of both using R-squared method. (Coefficient of determination)
aashrith48/SVM-on-geomechanical-data
Used support vector machine regressor algorithm to predict 2 labels from geomechanical data features. Multioutput regressor was used to predict 2 labels instead of a single prediction
aashrith48/K-Nearest-neighbours-with-HR-dataset
Used K- nearest neighbours classifier with HR dataset as features to predict the chances of an employee quitting, and compared it with logistic regression
aashrith48/Logistic-regression-on-HR-Data
Performed logistic regression on human resource data to for binary classification to predict whether an employee will quit a job or continue to stay.
aashrith48/K-means-clustering-on-geological-dataset
Performing K-means clustering on features consisting of geological data, along with optimal cluster number
aashrith48/Linear-regression-on-Geologic-Data
Performed linear regression on geologic data with appropriate visualization for viewing correlations between features to drop features for prediction
aashrith48/Hierachial-clustering-and-Isolation-forest-on-Petroleum-Data
Used hierarchial clustering methods to perform unsupervised learning on GIP-EUR data to classify into cluster.
aashrith48/Dimensionality-reduction-of-completions-data
Using Principal cluster analysis (PCA) and Non-negative matrix(NMF) for dimensionality reduction on a completions dataset consisting of
aashrith48/Shale-well-completions-visualization
Visualizing the completions data from a shale gas well using Plotly
aashrith48/Gas-Price
Visualizing Gas prices with matplotlib
aashrith48/Permeability-Prediction
Predicted reservoir permeability from well log data by obtaining optimal transformations through application of multiple regression (Alternating Conditional Expectation) algorithm
aashrith48/Production-Forecasting
Forecasted the fluid production from wells using historical time series data using time series analysis and Random Forest regression
aashrith48/Electrofacies-characterization-
This project involved characterizing the reservoir into different electrofacies based on well log data using Principal Component Analysis and Unsupervised Learning (K-Means clustering)