Fit a neural network based distribution model to data

This function delegates most work to keras::fit.keras.engine.training.Model() and performs additional consistency checks to make sure tf_compile_model() was called with the appropriate options to support fitting the observations y as well as automatically converting y to a n x 6 matrix needed by the compiled loss function.

Usage

# S3 method for reservr_keras_model
fit(
  object,
  x,
  y,
  batch_size = NULL,
  epochs = 10,
  verbose = getOption("keras.fit_verbose", default = 1),
  callbacks = NULL,
  view_metrics = getOption("keras.view_metrics", default = "auto"),
  validation_split = 0,
  validation_data = NULL,
  shuffle = TRUE,
  class_weight = NULL,
  sample_weight = NULL,
  initial_epoch = 0,
  steps_per_epoch = NULL,
  validation_steps = NULL,
  ...
)

Arguments

object: A compiled reservr_keras_model as obtained by tf_compile_model().
x: A list of input tensors (predictors)
y: A trunc_obs tibble of observed outcomes, or something convertible via as_trunc_obs().
batch_size: Integer or NULL. Number of samples per gradient update. If unspecified, batch_size will default to 32.
epochs: Number of epochs to train the model. Note that in conjunction with initial_epoch, epochs is to be understood as "final epoch". The model is not trained for a number of iterations given by epochs, but merely until the epoch of index epochs is reached.
verbose: Verbosity mode (0 = silent, 1 = progress bar, 2 = one line per epoch). Defaults to 1 in most contexts, 2 if in knitr render or running on a distributed training server.
callbacks: List of callbacks to be called during training.
view_metrics: View realtime plot of training metrics (by epoch). The default ("auto") will display the plot when running within RStudio, metrics were specified during model compile(), epochs > 1 and verbose > 0. Use the global keras.view_metrics option to establish a different default.
validation_split: Float between 0 and 1. Fraction of the training data to be used as validation data. The model will set apart this fraction of the training data, will not train on it, and will evaluate the loss and any model metrics on this data at the end of each epoch. The validation data is selected from the last samples in the x and y data provided, before shuffling.
validation_data: Data on which to evaluate the loss and any model metrics at the end of each epoch. The model will not be trained on this data. This could be a list (x_val, y_val) or a list (x_val, y_val, val_sample_weights). validation_data will override validation_split.
shuffle: shuffle: Logical (whether to shuffle the training data before each epoch) or string (for "batch"). "batch" is a special option for dealing with the limitations of HDF5 data; it shuffles in batch-sized chunks. Has no effect when steps_per_epoch is not NULL.
class_weight: Optional named list mapping indices (integers) to a weight (float) value, used for weighting the loss function (during training only). This can be useful to tell the model to "pay more attention" to samples from an under-represented class.
sample_weight: Optional array of the same length as x, containing weights to apply to the model's loss for each sample. In the case of temporal data, you can pass a 2D array with shape (samples, sequence_length), to apply a different weight to every timestep of every sample. In this case you should make sure to specify sample_weight_mode="temporal" in compile().
initial_epoch: Integer, Epoch at which to start training (useful for resuming a previous training run).
steps_per_epoch: Total number of steps (batches of samples) before declaring one epoch finished and starting the next epoch. When training with input tensors such as TensorFlow data tensors, the default NULL is equal to the number of samples in your dataset divided by the batch size, or 1 if that cannot be determined.
validation_steps: Only relevant if steps_per_epoch is specified. Total number of steps (batches of samples) to validate before stopping.
...: Unused. If old arguments are supplied, an error message will be raised informing how to fix the issue.

Value

A history object that contains all information collected during training. The model object will be updated in-place as a side-effect.

Details

Additionally, the default batch_size is min(nrow(y), 10000) instead of keras default of 32 because the latter is a very bad choice for fitting most distributions since the involved loss is much less stable than typical losses used in machine learning, leading to divergence for small batch sizes.

Examples

dist <- dist_exponential()
params <- list(rate = 1.0)
N <- 100L
rand_input <- runif(N)
x <- dist$sample(N, with_params = params)

if (interactive() && keras::is_keras_available()) {
  tf_in <- keras::layer_input(1L)
  mod <- tf_compile_model(
    inputs = list(tf_in),
    intermediate_output = tf_in,
    dist = dist,
    optimizer = keras::optimizer_adam(),
    censoring = FALSE,
    truncation = FALSE
  )

  tf_fit <- fit(
    object = mod,
    x = k_matrix(rand_input),
    y = x,
    epochs = 10L,
    callbacks = list(
      callback_debug_dist_gradients(mod, k_matrix(rand_input), x, keep_grads = TRUE)
    )
  )
}

Usage

Arguments

Value

Details

See also

Examples