MusiteDeep provides a deep-learning method for general and kinase-specific phosphorylation site prediction. It is implemented by deep learning library Keras and Theano backend (the Keras2.0 and Tensorflow backend implementation were also provided under folder MusiteDeep_Keras2.0). At present, MusiteDeep only provides prediction of human phosphorylation sites; however, it also provides customized model training that enables users to train other PTM prediction models by using their own training data sets based on either CPU or GPU.
Deep Learning for humans
In this repository, I will share some useful notes and references about deploying deep learning-based models in production.
Implementation of a Deep Learning chatbot using Keras with Tensorflow backend
R Package | Deep Learning | No Tensorflow or Theano backend | Variable Importance | Introducing Deeptree Algorithm
Flask API to productize a document classification model. Classification model was built using Keras with tensorflow backend
An implementation of resnet50 model from the paper "Deep residual learning for image recognition" by Kaiming He et.al in keras using tensorflow backend
A docker container that starts with fully featured Autonomio augmented intelligence workbench with Keras/Tensorflow deep learning backend.
Hyspeclib is a helper library written in python language on top of Tensorflow backend for analysis of hyperspectral images and perform classification task using supervised deep learning algorithm.