Table C.1.
AE imputer architecture construction.
Description | Size | Activation |
---|---|---|
Input | 10 | − |
Encoder | 7 | SELU |
5 | SELU | |
Decoder | 7 | SELU |
9 | SELU | |
Skip-connection | 19 | − |
Output | 9 | Linear |
Notes. Each row represents a layer with its size (number of neurons) and activation function. The input layer refers to the input set of magnitudes with missed values (grizy(u)JKSW1W2) and additional features (PSTAR in our case), and the output layer refers to the imputed magnitudes. The skip-connection concatenates the input and output of the decoder and has a size equal to 2× the number of magnitudes plus an additional input value (PSTAR).
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.