mx.callback.log: mx.callback.log

In MarkusBonsch/mxLSTM: LSTM model for regression

Description

custom logger function that is executed after each epoch of mxnet training.

Usage

mx.callback.log(period, loggerEnv)

Arguments

period	interval of logger execution in epochs.
loggerEnv	environment for storing the logger results

Value

The logger will create a list named 'logger' in loggerEnv with the following items:

• train numeric vector with training metric evaluations

• eval numeric vector with test metric evaluations

• checkpoint List of model weights for restoring the state after the specified epoch. See ?getMxNetModel on how to restore checkpoint

See Also

plot_trainHistory

MarkusBonsch/mxLSTM documentation built on May 28, 2019, 6:40 a.m.