Projects
Here is a list of Computer Vision projects I've developed primarily on Python along with its major deep learning frameworks and data manipulation libraries.
CG Texture Upscaler
Independent
Summary: The CG Texture Upscaler leverages my passion for computer vision and comuter graphics where I employ my understanding of deep learning to solve a fundamental industry problem: the lack of an appropriate tool to enhance outdated art assets, and sepcifically, textures, by upscaling them to higher resolutions through the use of AI. Existing industry leading toosl like Topaz and HitPaw do not handle computer graphics well and are not fit for automation.
​
The CG Texture Upscaler uses existing an ESRGAN model that is further trained on thusands of post-processed computer graphics that i've collected over the span of 5 months. I aimt to further develop a C-based version of the CG Texture Upscaler's CLI version.
​
Prominent technologies used: Python, PyTorch, OpenCV, ImageMagick.
​
​
Here is a list of end-to-end ML applications I've developed primarily on Python along with its major machine learning frameworks and data manipulation libraries.
Toronto Police MCI Crime Location Prediction
Independent
Summary: I developed this Python-based crime location prediction application that is accessible through its unique API through Heroku.
​
Prominent technologies used: Python, scikit-learn, Pandas/NumPy/PySpark, FastAPI, Docker.
​
​
Attention Seq-to-Seq English-Arabic Translation Machine
Independent
Summary: I developed this Python-based translation machine application that is accessible through it's unique API through Heroku and that uses attention seq-to-seq RNNs to handle sequence NLP data.
​
Prominent technologies used: Python, TensorFlow (with emphasis on Keras), Pandas/NumPy, FastAPI, Docker.
​
TFRS Book Recommender System
Independent
Summary: I developed this Python-based TFRS recommender system application that accessible through it's unique API and that features an embedding network trained on over 70 000 books and over 700 000 anonymized user data that I collected
​
Prominent technologies used: Python, TensorFlow Recommenders, Pandas/NumPy, Django, FastAPI, Docker.
​
BERT Review Sentiment Analyzer
Independent
Summary: I developed this Python-based review sentiment analyzer that features a BERT preprocessor and encoder, as well as a DNN classifier trained on 1 million+ comments that I collected and their respective ratings to predict anonymized reviewer sentiment.
​
Prominent technologies used: Python, TensorFlow and TF Hub, Pandas/NumPy, Django, FastAPI, Docker.
​
