/Deep-Learning

This repository records my learning journey in the realm of Deep Learning and its subfields.

Primary LanguageJupyter NotebookMIT LicenseMIT


Deep Learning

What we want is a machine that can learn from experience. — Alan Turing


About The Repository

This repository serves as an archive of my learning journey and projects that I have embarked in the realm of Deep Learning and its applications.

Feel free to contact me if you have any queries or spot any mistakes in my implementation.

Table of Contents

This section list out the projects in this repository.

Subfields Project Title Descriptions Keywords
CV EfficientNetV2 with Image Classification Benchmarks Utilizing SOTA model and training procedure on classic Image classification dataset like Fashion-MNIST and CIFAR10 using Timm running on Pytorch. Image Classification,
EfficientNetV2,
RandAugment,
Progressive Learning,
Timm,
Pytorch,
Faster-RCNN Vehicles Detection Detecting cars using Faster-RCNN with MobilenetV3 as backbone. Object Detection,
Faster-RCNN,
MobileNetV3,
FPN,
Timm,
Pytorch,
GAN AC-BIGGAN with CIFAR10 Generating small coloured images with AC-BIGGAN. ACGAN,
BIGGAN,
Conditional Batch Normalization,
Hinge Loss,
Label Smoothing,
IS,
FID,
Pytorch,
GDL_code Forked repository while reading O'Reilly's Generative Deep Learning book. VAE,
WGAN,
WGANGP,
CycleGAN,
Keras,
Generative-Adversarial-Networks-Projects Forked repository while reading PacktPublishing's Generative Adversarial Networks Projects book. 3DGAN,
cGAN,
DCGAN,
SRGAN,
StackGAN,
CycleGAN,
Keras,
RL Great Lunar Lander with DQN Solving Lunar-Landerv2 with DQNs approach. DQN,
DDQN,
SARSA,
OpenAI Gym,
Keras
Hands-On Reinforcement Learning with Python Coding exercise and self-made notes from the Hands-On Reinforcement Learning with Python book published by Packt. Reinforcement Learning,
Markov Decision Process,
Monte Carlo Control,
Otw,
Snake-DQN Solving Snake with DQN approach. DQN,
Keras

Prerequisites

The list of standard Python3 packages that I have used for my Machine Learning projects is shown in requirements.txt. To install all of the packages specific to each subfields, simply call the following command:

  • pip 
    cd CV
pip install -r requirements.txt

License

Distributed under the MIT License. See LICENSE for more information.

Contact

Zhao Wu Wong, Bryan - @LinkedIn - zhaowu.wong@gmail.com

Kaggle Profile: https://www.kaggle.com/kiritowu

Credits