farrokhkarimi / ai

Artificial Intelligence Fundamental Concepts

farrokhkarimi.github.io

ai

Artificial Intelligence Fundamental Concepts

Fundamental Concepts

Natural Neural Network and Artificial Neural Network Fundamental Concepts [3D Brain]

Image Processing

Image Processing with OpenCV [Colab notebook link]

TensorFlow and PyTorch Installation

TensorFlow installation guide
PyTorch installation guide
Google Colab

Deep Learning

• TensorFlow 2 quickstart
  • Built-in small datasets
  • Activation functions
  • Loss functions
  • Optimizers
  • Metrics
  • Callbacks
  • Architectures
• TensorFlow Datasets [Catalog]
• [Teachable Machine V1] [Teachable Machine V2]
• Convolutional Neural Network (CNN) [Paper] [Explainer]
• Image classification, [Data Augmentation], [Batch Normalization], [Overfitting], and [Dropout]
• Transfer learning and fine-tuning
• Autoencoders [Paper]
• Variational Autoencoder (VAE) [Paper]
• MusicVAE [Paper] [Reference]
• Generative Adversarial Network (GAN) [Paper1] [Paper2] [Paper3] [Scribble Diffusion] [ChatGPT] [GPTZero]
• Pix2Pix (Image-to-image translation with a [Conditional GAN]) [Paper1] [Paper2] [Reference] [Demo]
• Image Segmentation with U-Net [Paper]
• Recurrent Neural Networks (RNN) [Paper1] [Paper2] [Textbook]

Appendix

To plot the model:

tf.keras.utils.plot_model(model, rankdir="TD", show_shapes=True)

Parameters:
FC: (the previous layer number of nodes * the next layer number of nodes) + the next layer number of biases
CNNs: (kernel size (w*h) * number of channels * number of filters) + number of biases

Conv Feature Map Size:
The size of the convoluted matrix is given by C=((I-F+2P)/S)+1, where C is the size of the Convoluted matrix, I is the size of the input matrix, F is the size of the filter matrix and P is the padding applied to the input matrix.

Author

https://farrokhkarimi.github.io/