This repository is a comprehensive guide for learning PyTorch, from the basics like tensors to advanced topics like neural networks. Here you will find code examples, clear and concise explanations, and practical exercises to reinforce your PyTorch skills.
Before starting with this tutorial, it is recommended to have a basic understanding of Python and familiarity with machine learning concepts and neural networks. Additionally, you will need to have PyTorch installed in your development environment. You can find detailed instructions on how to install PyTorch in the official PyTorch documentation.
The repository is organized into different folders, each focusing on a specific topic. Below is a brief description of each folder
-
01_Tensors: This folder contains examples and exercises to understand the basics of tensors in PyTorch, including tensor creation, mathematical operations, dimension manipulation, and more.
-
02_Gradients: Here you will learn about automatic gradient computation in PyTorch and how to use them for parameter optimization in machine learning algorithms.
-
03_Simple_Models: In this folder, you will find examples of implementing simple machine learning models using PyTorch, such as linear regression and linear classifiers.
-
04_Neural_Network: Focusing more on neural networks, this folder covers the implementation of multi-layer neural networks (MLP) in PyTorch, including architecture definition, activation functions, training, and evaluation.
-
05_Convolutional_Neural_Networks: Here you will learn how to build and train convolutional neural networks (CNN) using PyTorch, one of the pillars of image processing and pattern recognition.
-
06_Recurrent__Neural_Networks: In the recurrent neural network (RNN) folder, you will dive into the world of time sequences and learn how to build and train RNN models using PyTorch.
-
07_Generative_Adversarial_Networks: Finally, in this folder, you will explore Generative Adversarial Networks (GAN), a special type of neural network architecture used for generating realistic synthetic data.
Each folder contains example code, theoretical explanations, and exercises that will allow you to delve deeper into each topic. You can follow the folder sequence to learn PyTorch from scratch or jump into the folder that interests you the most based on your needs.
Feel free to contribute to this repository with additional examples, corrections, or improvements. Any contribution is welcome. Additionally, this repository is distributed under the MIT License, which means you can use the code and resources available here for any purpose, both personal and commercial.
We hope you enjoy learning PyTorch and that this repository helps you master this powerful machine learning library!