- Given functional data
, first use Fourier basis functions to extract projection scores
by integration.
- J: number of projection scores for network inputs
- L: number of layers
- p: neurons per layer (uniform for all layers)
- s: dropout rate
- Loss function: softmax loss
- Batch size: data dependent
- Epoch number: data dependent
- Activation function: ReLU
- Optimizer: Adam
- "mfdnn_1d_par.R": hyperparameter selection with training data. More details can be found in comments
- "mfdnn_1d.R": functional deep neural netowrks. More details can be found in comments
- "mfdnn_2d_par.R": hyperparameter selection with training data. More details can be found in comments
- "mfdnn_2d.R": functional deep neural netowrks. More details can be found in comments
- "mfdnn_3d_par.R": hyperparameter selection with training data. More details can be found in comments
- "mfdnn_3d.R": functional deep neural netowrks. More details can be found in comments
- "example_1d.R": simulated data for one-dimensional functional data
- "example_2d.R": simulated data for two-dimensional functional data