AI Sketchpad

Implementations of ResNet-50 Deep Learning architecture.

Numpy Implementations

Neural networks implemented from scratch in numpy.

Keras Implementations

Neural networks implemented in keras.layers API

• Multi-Layer Perceptrons:
  • ResNet-50 in Keras Layers API - classification on RoCoLe + BRACOL dataset
  • ResNet-50 Transfer Learning - classification on RoCoLe + BRACOL dataset

Datasets

This section included dataset preprocessing notebooks. These need to be run first before corresponding neural network notebooks.

• Image datasets
  • RoCoLe
  • BRACOL

Debugging Techniques

Debugging techniques. Track input/output distributions, individual neuron weights, gradients, preactivation histograms,

Note this notebook has cool graphs, but no description

• MLP with Sigmoid Activation