---

Machine Learning Collection

In this repository you will find tutorials and projects related to Machine Learning. I try to make the code as clear as possible, and the goal is be to used as a learning resource and a way to lookup problems to solve specific problems. For most I have also done video explanations on YouTube if you want a walkthrough for the code. If you got any questions or suggestions for future videos I prefer if you ask it on YouTube. This repository is contribution friendly, so if you feel you want to add something then I'd happily merge a PR 😃

Table Of Contents

• Machine Learning Algorithms
• PyTorch Tutorials
  • Basics
  • More Advanced
  • Object Detection
  • Generative Adversarial Networks
  • Architectures
  • Lightning
• TensorFlow Tutorials
  • Beginner Tutorials
  • Architectures
• Docker setup

Machine Learning

•   Linear Regression - With Gradient Descent ✅
•   Linear Regression - With Normal Equation ✅
•   Logistic Regression
•   Naive Bayes - Gaussian Naive Bayes
•   K-nearest neighbors
•   K-means clustering
•   Support Vector Machine - Using CVXOPT
•   Neural Network

PyTorch Tutorials

If you have any specific video suggestion please make a comment on YouTube :)

Basics

•   Tensor Basics
•   Feedforward Neural Network
•   Convolutional Neural Network
•   Recurrent Neural Network
•   Bidirectional Recurrent Neural Network
•   Loading and saving model
•   Custom Dataset (Images)
•   Custom Dataset (Text)
•   Mixed Precision Training
•   Imbalanced dataset
•   Transfer Learning and finetuning
•   Data augmentation using Torchvision
•   Data augmentation using Albumentations
•   TensorBoard Example
•   Calculate Mean and STD of Images
•   Simple Progress bar
•   Deterministic Behavior
•   Learning Rate Scheduler
•   Initialization of weights

More Advanced

•   Text Generating LSTM
•   Semantic Segmentation w. U-NET
•   Image Captioning
•   Neural Style Transfer
•   Torchtext [1] Torchtext [2] Torchtext [3]
•   Seq2Seq - Sequence to Sequence (LSTM)
•   Seq2Seq + Attention - Sequence to Sequence with Attention (LSTM)
•   Seq2Seq Transformers - Sequence to Sequence with Transformers
•   Transformers from scratch - Attention Is All You Need

Object Detection

Object Detection Playlist

•   Intersection over Union
•   Non-Max Suppression
•   Mean Average Precision
•   YOLOv1 from scratch
•   YOLOv3 from scratch

Generative Adversarial Networks

GAN Playlist

•   Simple FC GAN
•   DCGAN
•   WGAN
•   WGAN-GP
•   Pix2Pix
•   CycleGAN
•   ProGAN
• SRGAN
• ESRGAN
• StyleGAN - NOTE: NOT DONE

Architectures

•   LeNet5 - CNN architecture
•   VGG - CNN architecture
•   Inception v1 - CNN architecture
•   ResNet - CNN architecture
•   EfficientNet - CNN architecture

PyTorch Lightning

•   Tutorial 1 - Introduction and starter code
•   Tutorial 2 - LightningModule
•   Tutorial 3 - Trainer
•   Tutorial 4 - Metrics
•   Tutorial 5 - DataModule
•   Tutorial 6 - Code restructure
•   Tutorial 7 - Callbacks
•   Tutorial 8 - TensorBoard logging
•   Tutorial 9 - Profiler
•   Tutorial 10 - Multi-GPU

TensorFlow Tutorials

If you have any specific video suggestion please make a comment on YouTube :)

Beginner Tutorials

•   Tutorial 1 - Installation, Video Only
•   Tutorial 2 - Tensor Basics
•   Tutorial 3 - Neural Network
•   Tutorial 4 - Convolutional Neural Network
•   Tutorial 5 - Regularization
•   Tutorial 6 - RNN, GRU, LSTM
•   Tutorial 7 - Functional API
•   Tutorial 8 - Keras Subclassing
•   Tutorial 9 - Custom Layers
•   Tutorial 10 - Saving and Loading Models
•   Tutorial 11 - Transfer Learning
•   Tutorial 12 - TensorFlow Datasets
•   Tutorial 13 - Data Augmentation
•   Tutorial 14 - Callbacks
•   Tutorial 15 - Custom model.fit
•   Tutorial 16 - Custom Loops
•   Tutorial 17 - TensorBoard
•   Tutorial 18 - Custom Dataset Images
•   Tutorial 19 - Custom Dataset Text
•   Tutorial 20 - Classifying Skin Cancer - Beginner Project Example

Docker Setup

Step 1: Install Docker

If you don't have Docker installed, follow the links below to install Docker for your system:

• Install Docker Engine

Step 2: Install Nvidia Container Toolkit (Optional)

If you plan to use GPU acceleration with CUDA, install Nvidia Container Toolkit:

• Nvidia Container Toolkit Installation Guide

Step 3: Build the Docker Image

Navigate to the directory containing the Dockerfile and build the Docker image with:

docker build -t aladdin-image .

Step 4: Run the Docker Container in Detached Mode

Run the following command to start the container in detached mode:

docker run -d \
  --gpus all \
  -v "${PWD}:/code" \
  -p 8080:8080 \
  --name "aladdin-container" \
  --env AUTHENTICATE_VIA_JUPYTER="mytoken" \
  aladdin-image \
  tail -f /dev/null

This will start a new Docker container named aladdin-container that will not exit immediately. The -d flag runs the container in detached mode, letting it run in the background.

Step 5: Interact with the Docker Container

To attach an interactive shell to the running container, use the command:

docker exec -it aladdin-container /bin/bash

You can now interact with your container using the bash shell.

Additional Notes

• If you wish to stop the container, you can do so with docker stop aladdin-container.
• To start the container again after stopping, use docker start aladdin-container.
• In case you need to remove the container, make sure it's stopped and then run docker rm aladdin-container.
• To see the output from the container (logs), use docker logs aladdin-container.