Deep Learning Notebooks

Deep Learning with Python

This repository follows Deep Learning with Python, Second Edition and Deep Learning with Python, Third Edition.

Getting Started

On Device, VPS, and More

When I first created this repo, I managed dependencies the old-fashioned way using %pip. After taking a 10-month break from the project, I recently returned to continue development—only to find that my environment had completely changed and nothing worked anymore.

I spent far too long juggling between pip, venv, poetry, and conda. Enough is enough. After some research, I found one tool that satisfied all my requirements: Pixi.

To get started, follow the Pixi installation guide.
Each folder in this repo is its own Pixi workspace, so you need to cd into the folder you want to work in before running commands.

To install dependencies, run:

pixi install

Each folder in this repo contains a pixi.toml, this controls pixi workspace. By default, most of workspace only have platform osx-arm64 enabled. If you are on a different platform, see this guide to enable your platform

Make sure you set the correct interpreter for your project. For VSCode, follow this guide.

Google Colab

If you want to run these notebooks on Google Colab. Simply fork this repo, go to File -> Open notebook -> Github and paste url to your fork and select the proper notebook. Colab come with most of dependencies installed so you shouldn't be needing to install yourself.

Or you can use colab's terminal and clone the repository.

The Table of Contents

1. MNIST
2. KerasTensorFlow
3. Generalization
4. KerasDeepDive
5. ComputerVision
6. ConvnetPatterns
7. ImageSegmentation
8. InterpetConvnet (Mostly broken)
9. ImageSgementation
10. ObjectDetection
11. TimeSeries

Keywords

• Tensorflow
• Keras
• Feedforward
• Convlution Neural Network
• Residual Neural Network
• Recurrent Neural Network
• Long Short-Term Memory
• Pretrain
• RetainNet
• MNIST
• Oxford-IIIT
• Yolo