Machine and Deep Learning Projects
Predicting House Value - NN Regression Project, Medical MRI Scans - CNN Computer Vision Project
Here are both my ML/DL projects.
The first was creating a classic Neural Network with PyTorch to predict the value of a house in San Francisco, based on its input data. This involved data-preprocessing, constructing the NN in PyTorch and evaluating hyper parameters such as network shape and gradient descent algorithm via k-cross-folds hyperparameter tuning.
The second project was a computer vision task and required implementing a U-Net Convolutional Neural Network architecture in PyTorch, to automatically detect potential cancerous regions in brain MRI scan. This project focused more on implementing the convolutional pass-through and feedback architecture, along with using performance metrics to accurately evaluate the quality of the model.
