Deep Learning from Scratch

This repository contains the coding practice for the book Deep Learning from Scratch. The main focus of this repository is to use unit tests extensively for the methods chapter by chapter. This structure allows the repository to serve as a self-paced coding exercise for the book.

Running Automated Tests

To run all the tests in the repository, use the following command:

pytest -s .

Notice: so many test is about trainning a neural network, which may costs a long time. Initially, all tests will fail. As you implement more methods, more tests should pass. This approach ensures that your progress is aligned with the book's content.

To run a specific test, you can use the -k option to target the desired test. For example:

pytest -s ch02_perceptron/test_perceptron.py -k test_AND

This command runs the test_AND test in the test_perceptron.py file located in the ch02_perceptron directory.

Code Quality and Checking

To maintain high code quality and consistency, this repository integrates several tools with pre-commit hooks. You can ignore this part if you don't want to do it.

Setting Up Pre-Commit Hooks

To set up the pre-commit hooks, run the following command:

pre-commit install

This installs the pre-commit hooks defined in the .pre-commit-config.yaml file. Once installed, these hooks will automatically check your code for any issues every time you make a commit.

Running Pre-Commit Hooks Manually

You can also manually run the pre-commit hooks on all files using the following command:

pre-commit run --all-files

This setup helps ensure that your code adheres to defined standards and maintains high quality throughout the development process.

---

Hope this repo can help you studying the deep learning.