All notebooks come in at least two versions:
lazy
: all code is ready to execute, doesn't contain output; andcomplete
: intended to be viewed on Github, does contain output.
For some notebooks, there is a courageous
version as well, which means
that you will have to write some of the code yourself.
- 010_underfitting_overfitting: illustrates the concepts of underfitting and and overfitting using non-linear regression.
- 020_mnist_data_exploration_complete: exploring the MNIST data set (complete notebook with output). 1 030_activation_function: visualization of the relevant activation functions.
- 040_mnist_mlp: illustrates data pipelines and construction of a classic multilayer perceptron to recognize handwritten digits.
- 050_convolution: illustrates convolution as used in convolutional neural networks.
- 060_mnist_cnn: illustrates a convolutional neural network to recognize handwritten digits.
- 070_imdb_data_exploration: exploring the IMDB data set.
- 080_imdb_rnn: illustrates recurrent neural networks using GRU and LSTM units for sentiment classification on the IMDB data set.