Explainable ML Vision
Developed an SVM Model for a Computer Vision classification task with explainable results, highlighting regions of importance.
Technologies Used:
Python Computer Vision SVM Explainable AI Feature Engineering HOG GridSearchCV scikit-learn
Wine Quality Classification
Trained GradientBoostClassifier, Random Forest, & Decision Tree models from the the scikit learn library. Tuned hyperparameters by using GridSearchCV.
Technologies Used:
Python scikit-learn Classification GradientBoostClassifier Random Forest Decision Tree GridSearchCV
Statistical Modeling with R
Various projects from my statistical modeling courses at UCI using the R programming language.
Technologies Used:
R LDA Multilevel Modeling Poisson Regression Nonlinear Models Principle Components Linear Regression
Keras Explainable Convolutional Neural Networks
Utilize the Keras GradientTape API to create Class Activation Maps (CAMs) that explain which regions deep Convolutional Neural Networks are focusing on for classification tasks.
Technologies Used:
Python Keras TensorFlow GradientTape CNN Class Activation Maps cv2 Image Interpolation
ML Models & functions from Scratch.
Multiple ML projects from my UCI courses which demonstrate understanding the fundamentals of popular ML models.
Technologies Used:
Python K Nearest Neighbors Linear Regression PyTorch Logistic Regression Decision Trees PCA CNN Clustering
Generative NLP: LSTM vs Markov Chains
Build the Markov Chains in Python using the given nusery rhymes dataset for training to create new nusery rhymes. Do the same with an LSTM model from Keras / TensorFlow, and compare results.
Technologies Used:
LSTM Markov Chains Generative Model NLP Language Processing