I am a graduate student at University of Southern California in Electrical and Computer Engineering specializing in Machine Learning, Deep Learning and Computer Vision.
Seeking co-op and full time opportunities.
I truly believe that creativity in the workplace, patience and staying long enough with the problem is the key to success.
Skilled in Python, C++,MATLAB, TensorFlow, Keras, OpenCV, Statsmodels, Sklearn, HTML, CSS, MySQL
-
Experience
1.Currently working on a project ‘Digital Pathology slides: Quality Control’:
Developing deep learning and machine learning based approaches by forming well-annotated and relatively artifact-free images to learn underlying disease-specific representations. Frameworks: PyTorch, cv2, OpenSlide, PIL, Sklearn, Torchvision, Docker containers.
2.Worked at Sat-Com Communications Solutions, Namibia:
Developed a D2D communication system between the CC3200 embedded WiFi IoT device and WiFi enabled radio devices.
Developed Digital Signal Processing based System for Voice Activity Detection and Background Noise Reduction.
3.Worked at Innovation Design Lab at Namibia University of Science and Technology as a Research Intern :
Contributed to Designing and Prototyping the Smart Power Management System for Solar Utility Vehicle.
-
Projects
1.Analysis of Tree based methods on APS failure dataset using SMOTE:
Trained a Logistic Model Tree using the pre-processed data for classification of data on the APS Failure dataset.
2.Analysis of machine learning models on the communities and crime dataset:
Implemented a Linear Model, Ridge Regression Model, Lasso Regression Model, PCR model, L-1 penalized Gradient boosting tree using XGBoost.
3.Breast-Cancer-Analysis:
Developed a Supervised Learning (L-1 penalized SVM), Semi-Supervised Learning/ Self-training: Unsupervised Learning (k-means and Spectral Clustering).
4.Activity-Recognition-system (Time series classification):
Human activities are classified based on time series obtained by a Wireless Sensor Network.
Binary Classification (Logistic Regression), Multi-class Classification (Multinomial Logistic Regression and Gaussian Naïve Bayes)
5.Generative Models for Text (LSTM training):
Developed a generative model to mimic the writing style of prominent writers.
6.(Deep) CNNs for Image Colorization:
Developed a convolutional neural network for image colorization which turns a grayscale image to a colored image
7.Developed a Pattern Recognition System on the real-world data set from UCI ML Archive
and comparison is made between different types of supervised algorithms.
8.Image processing: Image Classification of CIFAR-10 data set using Convolutional Neural Networks (CNN):
Comparison of Lenet -5 CNN and Successive Subspace Learning CNN is implemented on basis of accuracy and model size.
9.Image Processing: Texture Analysis and segmentation is done on the images which involves feature extraction, feature averaging, feature reduction and both supervised and unsupervised classification techniques.
Applications of Morphological processing, shape analysis and retrieval is also implemented.
10.Email Classification, Machine Learning:
Implemented and contrasted the algorithms: K Nearest Neighbors, Support Vector Machine, Naive Bayes Classifier.
Employed multiple libraries (Sklearn, NumPy, Matplotlib) and algorithms (TF-IDF, LSA) to classify emails.
11.License Plate Reader, Pattern Recognition:
Automated number plate recognition using Convoluted Neural Networks and other image processing techniques.