Datasets

Text classification

• Amazon reviews: as used for the first time in Blitzer’s 2007 paper
• Amazon reviews idf-tf: Amazon reviews after preprocessing done in Chen’s 2011 paper, reducing dimensionality and applying idf-tf re-weighting. The link is at the end of the README file.

  Digit classification

• MNIST
• USPS: link1, link2
• SVHN
• MNIST-M: allows building an MNISTM dataset from the MNIST and the BSDS500 dataset

Dataset repositories

• UCI ML repository
• mldata: This repository is in particular used by Scikit-Learn to download datasets via the datasets.fetch_mldata function.
• OpenML: This repository is in particular used by Scikit-Learn to download datasets via the datasets.fetch_openml function.
• libsvm library: in addition to the library files, contains several datasets.
• Repository for transfer learning datasets

Tools

• Google PAIR tools: GUI ML tools.

  Latex

• Cleaning .bib files

  Visualization and graphing

• Mermaid live editor: make Gantt diagrams, flow charts etc… in a fast way.

  Deep Learning

• TensorSpace.js: 3D visualization of deep neural networks
• 3D visualization of CNN’s: a 3D visualization of a CNN used to classifiy digits, with the possibility to test on the user’s own hand writing.

Code links

Domain adaptation

• TensorFlow (old) and Keras implementation of some Deep Domain Adaptation Networks
• PyTorch implementation of some Deep Domain Adaptation Networks
• salad: a Domain adaptation library written in PyTorch

  Paper-code association

• Papers with code
• ResearchCode

Miscellaneous

• Awesome transfer learning repository: with DA datasets
• Distill: interactive ML articles
• A github repository for transfer learning
• Nevergrad: A gradient-free optimization platform, by facebook research
• Online library and publication platform: several pdf books available for download.