powered by
Multiplies the weights by (1 - weightDecay) before applying the scheduled weight changes.
weightDecay
weightDecayWeightUpdate(darch, layerIndex, weightsInc, biasesInc, ..., weightDecay = getParameter(".darch.weightDecay", 0, darch), debug = getParameter(".debug", F, darch))
'>DArch instance.
Layer index within the network.
Matrix containing scheduled weight updates from the fine-tuning algorithm.
Bias weight updates.
Additional parameters, not used.
Weights are multiplied by (1 - weightDecay) before each update. Corresponds to the darch.weightDecay parameter of darch.default.
darch.weightDecay
Internal debugging flag.
updated weights
Other weight update functions: maxoutWeightUpdate
maxoutWeightUpdate
# NOT RUN { model <- darch(Species ~ ., iris, c(0, 50, 0), darch.weightUpdateFunction = "weightDecayWeightUpdate") # }
Run the code above in your browser using DataLab