Open Access

Fig. 4

image

Download original image

Chosen architecture of the neural network (NN) to predict single zone model's SEDs. The NN consists of an input layer (orange) with 18 neurons (18 free parameters), 6 hidden layers (blue circles) with 300 neurons each, and an output layer (red) with 140 neurons (the SED values at wavelength i of the 140 points (SED(λi)). All neurons are connected to every neuron in the adjacent layer. This sums up to 499 340 trainable parameters.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.