Chainer modules

Comparing with caffe

If you are using caffe, it is easy to get accustomed to chainer modules. Several variable’s functionality is similar and you can see below table for its correspondence.

datasetsData layersInput data can be formatted to this class for the model input.
It covers most of the use case of input data structure.
variableblobIt is an input and output of functions/links/Chain.
functionslayersFramework supports widely used functions in deep learning. Ex sigmoid, tanh, ReLU etc.
linkslayersFramework supports widely used layers in deep learning. Ex Linear layer, Convolutional layer etc.
Chainnetlinks and functions (layers) are jointed to form a ”model”.
This group of layers are Chain/net
optimizerssolverSpecify what kind of gradient descent method to be used for tuning model parameter. Ex. SGD, AdaGrad, Adam.
serializers Save/load training state. Ex. model, optimizer etc.
iterators – Defines each minibatch construction, used in Trainer. 
training.updater – Defines each forward-backward propagation followed by parameter update procedure, used in Trainer.
training.Trainer – It manages training procedure.


VariableFunctionLink and Chain is explained in Chainer basic module introduction.

Optimizer and Serializer is explained in Chainer basic module introduction 2.

Leave a Comment

Your email address will not be published. Required fields are marked *