Reference sparse coding implementations for efficient learning and inference implemented in PyTorch with GPU support.
- Repo currently includes classic patch-wise sparse coding dictionary learning.
- Locally Competative Algorithm (LCA)
- Gradient Descent with Euler's method on Laplace Prior (Vanilla)
- Laplacian Scale Mixture (LSM)
- Iterative Shrinkage-threshold Algorithm (ISTA)
- Generic PyTorch minimization of arbitrary loss function (PyTorchOptimizer)
- Clone the repo.
- Navigate to the directory containing the repo directory.
- Run
pip install -e sparsecoding - Navigate into the repo and install the requirements using
pip install -r requirements.txt - Install the natural images dataset from this link: https://rctn.org/bruno/sparsenet/IMAGES.mat
- Try running the demo notebook:
examples/sparse_coding.ipynb
Note: If you are using a Jupyter notebook and change a source file, you can either: 1) restart the Jupyter kernel, or 2) follow instructions here.
See the contributing document!