This page provide essential resources about PyTorch.
(PyTorch is better for developing research code — it is more intuitive and has much better document and communities. Also, it is much more easier comparing to TensorFlow. I have used TensorFlow quite a bit.)
(PyTorch compiles faster, FAIR does many interesting work, and share their code a lot earlier than Google. In turn, many modules have diverse implementation in pytorch.)
Thanks Subarna and Jiasen for their very helpful help.
PyTorch is an open source machine learning library based on the Torch library, used for applications such as computer vision and natural language processing, primarily developed by Facebook‘s AI Research lab (FAIR). It is free and open-source software released under the Modified BSD license. Although the Python interface is more polished and the primary focus of development, PyTorch also has a C++ interface.
PyTorch provides two high-level features:
It’s a Python-based scientific computing package targeted at two sets of audiences:
- A replacement for NumPy to use the power of GPUs
- a deep learning research platform that provides maximum flexibility and speed
PyTorch tutorials for Beginners
Deep Learning with PyTorch: A 60 Minute Blitz
Make sure you have the torch and torchvision packages installed.
PyTorch in 5 Minutes (Code for this video at here)
[warning, just watch to get a big picture of PyTorch, do not actually do it as the video is published on Apr 30, 2017, things in PyTorch doc can change and some function can become deprecated]
From comments of this video, “PyTorch is for people who do research and experimentation. TF (and Keras) are for people who just dabble in DL and just need to get the stuff done asap. Frankly, I don’t know any researcher who actually uses TF.“
Faster R-CNN Object Detection with PyTorch (JUNE 18, 2019)
Building your own object detector — PyTorch vs TensorFlow and how to even get started? (Apr 25, 2020) [PDFif not retrievable]
To understand more about how to use PyTorch for your deep learning project, I recommend checking this great book: Deep Learning with PyTorch, one of the best resources to learn about PyTorch.
PyTorch Geometric (PyG) is a geometric deep learning extension library for PyTorch.
PyTorch3D is FAIR’s library of reusable components for deep learning with 3D data https://pytorch3d.org/