Introduction:

A repository dedicated to implement many sorts of algorithms in Machine Learning, Deep Learning and Reinforcement Learning using TensorFlow 2 (emphasising heavily in this library) and many other libraries.

Algorithms:

[x] Machine Learning:

• Linear Regression
• Kernel Linear Regression
• Logistic Regression (0,1)
• Logistic Regression (-1,1)
• Kernel Logistic Regression
• K Nearest Neighbours
• K Means Clustering
• Naive Bayes Classifier
• Linear Support Vector Machine
• Non Linear Support Vector Machine

[x] Computer Vision:

• Auto-encoders
  • Vanilla Autoencoder
  • Convolutional Autoencoder
  • Sparse Autoencoder
  • Stacked Autoencoder
  • Denoise Autoencoder
  • Variational Autoencoder
  • Contractive Autoencoder
  • Adversarial Autoencoder
  • Self-attention Autoencoder
  • Recurrent Autoencoder
  • Sparse Variational Autoencoder
  • Beta Variational Autoencoder
  • Transformative Autoencoder
• Generative Adversarial Network
  • Deep ANN GAN
  • Deep Convolutional GAN
  • Conditional GAN
  • Least Square GAN
  • Wasserstein GAN
  • Wasserstein Improved GAN
  • Deep Regret Analytic GAN
  • Cycle GAN
  • Star GAN
  • Super Resolution GAN
• Image Classification
  • Large Network
    • AlexNet
    • VGG Net
    • Inception Net
      • Inception V1 (Google Net)
      • Inception V2
      • Inception V3
      • Inception V4
    • Residual Net
    • Res Next
    • SeNet
    • Dense Net
    • Highway Net
    • Network In Network
    • Vision Transformer
    • Residual Attention Net
    • Capsule Network
  • Small Network
    • LeNet
    • ZFNet
    • Deep Compression
    • Knowledge Distillation
    • MLP-Mixer
    • Mobile Net
    • Poly Net
    • Shuffle Net
    • Squeeze Net
    • Xception Net
    • XNOR-Net
  • Robustness
    • Adversarial Saliency Maps
    • Black Box Methods
    • Fast Gradient Sign Method
    • Iterative Least Likely Method
• Miscellaneous
  • Genetic CNN
  • Style Transfer
  • Empirical Risk Minimisation
  • Gaussian CDF
  • Gradient Accumulation
  • Gradient test for Initialisation
  • Regularisation with Nelder Mead Optimisation

[x] Reinforcement Learning:

• Memory:
  • Replay Buffer
  • Non Buffer
  • Prioritised Experience Replay
• Strategies:
  • Value Based:
    • DQN
    • Double DQN
    • Duelling DQN
    • Double Duelling DQN
    • Rainbow
    • C51
  • Policy Based:
    • PG (vanilla)
    • Reinforce
    • Trusted Region PO (TRPO)
    • Proximal PO (PPO)
    • ACKTR
  • Actor Critic Based:
    • Vanilla AC
    • Advantageous AC (A2C)
    • Asynchronous Advanatageous AC (A3C)
    • Soft AC (SAC)
    • Deep Deterministic PG (DDPG)
    • Twin Delayed DDPG (TD3)

Contributions

Contributions are welcome! If you have implemented a new insights or have improvements to existing implementations, feel free to submit a pull request. Please follow the contribution guidelines outlined in the repository.

License

This project is licensed under the GNU General Public License v3.0 - see the LICENSE file for details.

Happy generating!