Name: Christopher Erick Moody
Type: User
Company: @gumtap
Bio: Work on Bayesian Deep Learning, Factorization Machines, NLP, lda2vec, sklearn & Chainer Framework contributor.
Location: San Francisco
Blog: twitter.com/chrisemoody
Christopher Erick Moody's Projects
An artificial neural network that attempts to impersonate the writing style of any given text training set
Automatically exported from code.google.com/p/beta-ntf
BH T-SNE
A flexible framework of neural networks for deep learning
Collaborative filtering with the GP-LVM
A curated list of data science blogs
Putting the Unite the People project in a docker image
Finding document vectors from pre-trained word2vec word vectors
Personal Dotfiles
Tailwind-kit is a free and open source components kit fully coded with Tailwind css 2.0.
PyTorch Flexible Hash Embeddings
Gaussian Processes with kernels specific to business patterns (day of week, holiday, yearly and seasonal calendars)
Project page for End-to-end Recovery of Human Shape and Pose
Easily turn large sets of image urls to an image dataset. Can download, resize and package 100M urls in 20h on one machine.
IPython Notebook Theme. Base16 default colors, VIM mode by default
AWS Lambda function with Pandas, Sklearn, Scipy, Numpy and Chainer
Alternative to Parquet. 100x faster for random access, includes a vector index and automatic versioning. Apache Arrow and DuckDB compatible.
The dataset contains 3 million attribute-value annotations across 1257 unique categories on 2.2 million cleaned Amazon product profiles. It is a large, multi-sourced, diverse dataset for product attribute extraction study.
Extending Random Forests for Vectors
Poincare Factorization Machines
Python library for interactive topic model visualization. Port of the R LDAvis package.
load caffe prototxt and weights directly in pytorch