projects

machine learning

• Supervised Learning [Code]
  • Classified MNIST dataset using MAP, MLE, Bayesian pairwise, and perpendicular bisectors to understand the fundamentals of data and classification.
  • Classified Face using six different features (Eigen Face, Fisher Face, KernelPCA, Kernel Fisher Face, VGG Face, ResNet) by training MLP classifier on Yale Face, Indian Movie Face, and IIT-CFW Database and used t-SNE for face visualization.
  • Implemented data dimensionality reduction using PCA and ISOMAP on the CIFAR-10 dataset and classified using SVM.
• Unsupervised Learning [Code]
  • Implemented Manifold learning methods - MDS, LLE, and ISOMAP. Performed k-means and spectral clustering on the Concentric Circles and Swiss roll dataset and visualized using manifold in 2-D.

speech processing

• Speech projects [Code]
  • Neural style transfer of audio(spectrogram) with PyTorch based on the algorithm developed by [Leon A. Gatys, Alexander S. Ecker and Matthias Bethge]
  • Understanding and Implementation of wavelet Transform and Mallat S Scattering Transform using [Kymatio: Scattering Transforms in Python]