Facebook Developers Resources Newsletters #6: Introduction to PyTorch P1
A. What is PyTorch
PyTorch is Open source deep learning research platform that provides maximum flexibility and speed. It has been developed by Facebook‘s artificial intelligence research group(FAIR), and Uber‘s Pyro software for probabilistic programming is built on it.
PyTorch is inspired by Torch and based on Python. As you know, Python now is the most popular programming language for Machine Learning Engineers, Data Scientists and AI researchers or folks who love Data.
The creators of PyTorch said that “Deep integration into Python allows popular libraries and packages to be used for easily writing neural network layers in Python.”
B. Philosophy of PyTorch’s development
- Stay out of the way
- Cater to the impatient
- Promote linear code-flow
- Full interop with the Python ecosystem
- Be as fast as anything else
C. Features:
1. HYBRID FRONT-END
PyTorch’s Hybrid front-end provides ease-of-use and flexibility in eager mode, while seamlessly transitioning to graph mode for speed, optimization, and functionality in C++ runtime environments.
FAIR
2.DISTRIBUTED TRAINING:
Optimize performance in both research and production by taking advantage of native support for asynchronous execution of collective operations and peer-to-peer communication that is accessible from Python and C++.
3. PYTHON-FIRST:
PyTorch is not a Python binding into a monolithic C++ framework. It’s built to be deeply integrated into Python so it can be used with popular libraries and packages such as Cython and Numba.
4. TOOLS & LIBRARIES
An active community of researchers and developers have built a rich ecosystem of tools and libraries for extending PyTorch and supporting development in areas from computer vision to reinforcement learning
Let’s get ready to move forward with deep learning and neural networks. In the next articles, we will get PyTorch installed. I’ll see you there!
References:
- PyTorch's website: https://pytorch.org/features
- Facebook AI: https://facebook.ai/developers/tools/pytorch