Data Science and Machine Learning Portfolio

• My LinkedIn
• Repository containing portfolio of data science projects completed for academic, self learning, and professional purposes.

Masters thesis

• Satellite-based detection of illegal manure application in farmland parcels: Supervisor: Prof. Dell'Acqua

Published Articles

• Gradient Descent: I explained how gradient descent works mathematically, and its varients to improve performance.
• Overfitting: My article about what is overfitting, how to detect it, and how to avoid it.
• Cost Functions: My article about cost functions. Covered cost functions include: MSE, RMSE, MAE, MBE, Cross Entropy.

Machine Learning:

• Unsupervised Learning / Clustering Iris Dataset: Implementation of K-Means Algorithm without any ready library functions.

Computer Vision

• Yolov3 Fine Tuning: Fine tuning the Yolov3 model with custom dataset to improve performance.
• Emotion Recognition: Basic CNN model and Transfer Learning Techniques to predict emotions of people from given images.
• Object Detection: Object Detection in sequences of frames.
• Dog Breed Classification: Implementation of a Dog breed classification with FastAi Library.
• UAV Detection: Detection of UAV's with YOLOv3.

Natural Language Processing

• Speech Recognition Flask App: Built a Flask App that gets a WAV file, and with Speech Recognition library, prints the transcripts.

Data Analysis

Certificates

• Machine Learning Crash Course by Google: Credential ID 948342822658
• Python for Data Science and AI

Course Work

• Data Visualization: Data Visualization on Wealth Nations Dataset.
• Statistical Analysis on EFSA dataset : Perfomed statistical analysis and feature selection on EFSA dataset for the Statistical Learning course in my Master's Degree.