Fit a deep learning model
Usage
tl_fit_deep(
data,
formula,
is_classification = FALSE,
hidden_layers = c(32, 16),
activation = "relu",
dropout = 0.2,
epochs = 30,
batch_size = 32,
validation_split = 0.2,
learning_rate = NULL,
verbose = 0,
...,
compute = "cpu"
)Arguments
- data
A data frame containing the training data
- formula
A formula specifying the model
- is_classification
Logical indicating if this is a classification problem
Vector of units in each hidden layer (default: c(32, 16))
- activation
Activation function for hidden layers (default: "relu")
- dropout
Dropout rate for regularization (default: 0.2)
- epochs
Number of training epochs (default: 30)
- batch_size
Batch size for training (default: 32)
- validation_split
Proportion of data for validation
- learning_rate
Optimizer learning rate. NULL (default) leaves keras's own adam default in place. (default: 0.2)
- verbose
Verbosity mode (0 = silent, 1 = progress bar, 2 = one line per epoch) (default: 0)
- ...
Additional arguments
- compute
Compute tier. Either
"cpu"(default) or"gpu". GPU usage is handled automatically by the underlying tensorflow runtime when CUDA is configured; this argument is accepted for API consistency with the rest of tidylearn but does not itself change the keras model setup. The expectation is that the caller has already resolved the compute tier viatl_compute_advisor/tl_resolve_compute.