Final Project - Software Engineering 20211
-
Updated
Feb 28, 2022 - Python
Final Project - Software Engineering 20211
A bitcoin price forecaster utilizing an ensemble of autoregressive, N-BEATS, LSTM and layer normalized models, trained on various loss functions.
PixelCNN implementation - based on the original version
A JAX implementation of DEformer for arbitrary conditioning.
repo for practicing DL/genAI
causal decoder based on convolutions only (no attention): can be applied to ubbounded sequence lengths; the prediction of the next token depends on *all* previous tokens; allows autoregressive sampling; highly gpu-parralellizable; trained with teacher forcing;
The code of GMM and MAF classifiers
Backtesting software for intraday and daily timeframe SARIMAX forecasting model. Capable of forecasting on any asset class with backtesting capability across various parameter sets customizable by the user.
Research on Image Compression using Deep Neural Networks
Source Separation of Multi-source Raw Music using Residual Quantization
[CVPR 2022] Official implemention of the paper "LAR-SR: A Local Autoregressive Model for Image Super-Resolution“
Light, modular framework for dynamic time series modeling, compatible with scikit-learn
Implementation of the genetic algorithm for structural break detection in time series that chooses a piecewise autoregressive model using minimum description length principle
Implementing Bayes by Backprop with PyTorch. Applied on time-series prediction.
Noise-conditional score networks for music composition by annealed Langevin dynamics
PixelPyramids: Exact Inference Models from Lossless Image Pyramids (ICCV 2021)
PyTorch Lightning Implementation of Diffusion, GAN, VAE, Flow models
Sequence-to-Sequence Generative Model for Sequential Recommender System
[ICML 2023] Architecture-Agnostic Masked Image Modeling -- From ViT back to CNN
Physics-informed convolutional-recurrent neural networks for solving spatiotemporal PDEs
Add a description, image, and links to the autoregressive topic page so that developers can more easily learn about it.
To associate your repository with the autoregressive topic, visit your repo's landing page and select "manage topics."