PyTorch tutorials for beginners

Introduction

What is PyTorch?

PyTorch is a machine learning framework based on the Torch library created by Soumith Chintala at Meta(Facebook).

PyTorch was initially used more in research for fast prototyping , but lately it is used more and more in production.

Why PyTorch over tensorflow? (personal opinion)

• Both PyTorch and tensorflow are very strong frameworks, they reduce a huge amount of work compared to other libraries or starting from scratch.

• The main difference is PyTorch is more pythonic than tensorflow, so if you're used to coding in python then PyTorch is your pick.

• in PyTorch you have access to low level code.

• When researchers want flexibility & ease debugging and short training duration, they choose PyTorch.

• Tensorflow beats pytorch in deployment, so if you aim to deploy your models into a mobile device then tensorflow is best.

• In the area of data parallelism, PyTorch gains optimal performance by relying on native support for asynchronous execution through Python. However, with TensorFlow, you must manually code and optimise every operation run on a specific device to allow distributed training.

Now that you are ready for PyTorch. Let's start directly with

1. Prerequisites

2. Preparing your deep learning environment

3. Installing python & pytorch

4. Basic python

5. Tenosrs & Operations

TODO