considering only those not directly falling into ML or DL category
considering only those not directly falling into DL category
Ian Goodfellow and Aaron Courville in their book titled “Deep Learning” define Deep Learning in terms of the depth of the architecture of the models:
The hierarchy of concepts allows the computer to learn complicated concepts by building them out of simpler ones. If we draw a graph showing how these concepts are built on top of each other, the graph is deep, with many layers. For this reason, we call this approach to AI deep learning.
The above is often simplified to:
Deep learning is the implementation of neural networks with more than a single hidden layer of neurons.
A key difference from ML is the ability to perform automatic feature extraction from raw data, also called feature learning.
nrow(available.packages())returned 14127 CRAN packages not counting Bioconductor repository, github, gitlab etc
Two major data science languages: Python and R
Scalable, Portable and Distributed Gradient Boosting (GBDT, GBRT or GBM) Library, for Python, R, Java, Scala, C++ and more. Runs on single machine, Hadoop, Spark, Flink and DataFlow https://xgboost.ai/ an implementation of gradient boosted decision trees designed for speed and performance
A fast, distributed, high performance gradient boosting (GBDT, GBRT, GBM or MART) framework based on decision tree algorithms, used for ranking, classification and other machine learning tasks. https://lightgbm.readthedocs.io/en/latest/
A fast, scalable, high performance Gradient Boosting on Decision Trees library, used for ranking, classification, regression and other machine learning tasks for Python, R, Java, C++. Supports computation on CPU and GPU. https://catboost.ai
H2O is an open source, in-memory, distributed, fast, and scalable machine learning and predictive analytics platform that allows you to build machine learning models on big data and provides easy productionalization of those models in an enterprise environment.
MXNet is a machine-learning framework with APIs is languages such as R, Python and Julia which has been adopted by AWS.
Keras is a high-level neural networks API developed with a focus on enabling fast experimentation. Being able to go from idea to result with the least possible delay is key to doing good research.
The R interface to TensorFlow lets you work productively using either high-level APIs or the core TensorFlow API.
NB: while TensorFlow models are typically defined and trained using R or Python code, it is possible to deploy TensorFlow models in a wide variety of environments without any runtime dependency on R or Python
interpretable-ml-book Making Black Box Models Explainable
These 4 CRAN packages are in active development:
Shapper is a part of the DALEX universe. The DALEX universe is a part of the DrWhy.AI universe, the collection of tools for Explainable AI (XAI). It’s based on shared principles and simple grammar for exploration, explanation and visualisation of predictive models.
CRAN task view ModelDeployment reviews R packages, grouped by topic, that provide functionalities to streamline the process of deploying models to various production environments.
Docker allows you to wrap up your R product in a self contained mini computer that can then be easily shared and run in a variety of different environments.
Docker file builds docker image which runs docker container.
The Rocker Project Docker Containers for the R Environment
Dockerfile for GPU enabled rstudio server with keras+tensorflow
to run the code from the Deep learning in R book given an NVIDIA GPU with drivers installed.