Pinned Repositories
3d_diffeomorphic_nnets
3D non-order specified LLDDM-resnet pseudo-diffeomorphic net implementation in Tensorflow.
baubels
Github Profile
baubels.github.io
Personal website
bd-covid19-dash-app
Repo for deploying data visualisation app for Bangladeshi Covid-19 data to Heroku
diffeomorphic_registration
Image registration for diffeomorphically comparable shapes via neural nets. (paper implementation, done with Tensorflow).
fcnns
Custom (paper) implementation for fourier neural nets - implementing max pooling in the fourier basis for increased efficiency.
gradient_teleportation
Improving neural net convergence speeds via post-activation invariant actions with respect to the loss function.
image_clustering
A simple unsupervised image clustering library. Offers Neural embeddings with a MobileNetV2, HOG feature descriptors, PCA/LLE/Spectral dimensionality reduction schemes, KMeans clustering, and a WSS dispersion-based cluster elbow test.
numpynets
Neural nets from scratch using NumPy.
reinforcement_learners
A mini RL library implementing RL learners on OpenAI environments
baubels's Repositories
baubels/diffeomorphic_registration
Image registration for diffeomorphically comparable shapes via neural nets. (paper implementation, done with Tensorflow).
baubels/image_clustering
A simple unsupervised image clustering library. Offers Neural embeddings with a MobileNetV2, HOG feature descriptors, PCA/LLE/Spectral dimensionality reduction schemes, KMeans clustering, and a WSS dispersion-based cluster elbow test.
baubels/3d_diffeomorphic_nnets
3D non-order specified LLDDM-resnet pseudo-diffeomorphic net implementation in Tensorflow.
baubels/baubels
Github Profile
baubels/baubels.github.io
Personal website
baubels/bd-covid19-dash-app
Repo for deploying data visualisation app for Bangladeshi Covid-19 data to Heroku
baubels/fcnns
Custom (paper) implementation for fourier neural nets - implementing max pooling in the fourier basis for increased efficiency.
baubels/gradient_teleportation
Improving neural net convergence speeds via post-activation invariant actions with respect to the loss function.
baubels/numpynets
Neural nets from scratch using NumPy.
baubels/reinforcement_learners
A mini RL library implementing RL learners on OpenAI environments
baubels/covid19-dashboard
🦠Django + Plotly Coronavirus dashboard. Powerful data driven Python web-app, with an awesome UI. Contributions welcomed! Featured on 🕶Awesome-list
baubels/fastbook
Draft of the fastai book
baubels/flt3
A one-page proof of Fermat's Last Theorem for the case n=3.
baubels/forward-forward
Hinton's forward forward discriminative model in Pytorch
baubels/hello-world
I try to re-remember how to use git
baubels/hnns
Harmonic Neural networks implementation in tensorflow
baubels/illinois-crop-yield
Portable corn crop yield predictor for farmers in Illinois. Purposefully designed to use inexpensive priors, with built-in functionality to use more.
baubels/image-dataset-generator
Algorithmically generated image dataset
baubels/inf_itgames
Prisoner Dilemma exposé, including dominant, noisy, isolatory, extortionate, and self-made strategies.
baubels/latency_champ
Microprocessor-aware time series stock predictor using deep learning. 'In-house' the quickest and most accurate stock predictor of the 2021 Algothon competition.
baubels/live_edge_detection
Performs edge detection live on user video input
baubels/no-style-please
A (nearly) no-CSS, fast, minimalist Jekyll theme.
baubels/real_number_game
My solutions to some version of the real number game
baubels/rings-1
I play with rings in the lean theorem prover
baubels/SAAFs
Applies structure adaptive filtering to an input image.
baubels/seven-dwarfs
Render Web app which recognises each of snow white's seven dwarfs from pictures using deep learning
baubels/stuod_cyclone
Geo-temporal tropical cyclone associator and automated regressor/classifier on US Naval Research Laboratory's ~2000s weather data.
baubels/tf_resnets
code for custom resnet18 implementation in tensorflow
baubels/timeseries_visualised
baubels/unsupervised_porto
A short (unsupervised) analysis of student (portuguese) background correlation with respect to grades.