Save/Load models using SavedModel format
save_model_tf(
object,
filepath,
overwrite = TRUE,
include_optimizer = TRUE,
signatures = NULL,
options = NULL
)load_model_tf(filepath, custom_objects = NULL, compile = TRUE)
Model object to save
File path
Overwrite existing file if necessary
If TRUE
, save optimizer's state.
Signatures to save with the SavedModel. Please see the signatures
argument in tf$saved_model$save
for details.
Optional tf$saved_model$SaveOptions
object that specifies options
for saving to SavedModel
Mapping class names (or function names) of custom (non-Keras) objects to class/functions (for example, custom metrics or custom loss functions). This mapping can be done with the dict() function of reticulate.
Whether to compile the model after loading.
Other model persistence:
get_weights()
,
model_to_json()
,
model_to_yaml()
,
save_model_hdf5()
,
save_model_weights_hdf5()
,
serialize_model()