tf.train.warm_start( ckpt_to_initialize_from, vars_to_warm_start='.*', var_name_to_vocab_info=None, var_name_to_prev_var_name=None )
Warm-starts a model using the given settings.
If you are using a tf.estimator.Estimator, this will automatically be called during training.
ckpt_to_initialize_from: [Required] A string specifying the directory with checkpoint file(s) or path to checkpoint from which to warm-start the model parameters.
vars_to_warm_start: [Optional] A regular expression that captures which variables to warm-start (see tf.get_collection). Defaults to
'.*', which warm-starts all variables. If
Noneis explicitly given, only variables specified in
var_name_to_vocab_infowill be warm-started.
var_name_to_vocab_info: [Optional] Dict of variable names (strings) to VocabInfo. The variable names should be "full" variables, not the names of the partitions. If not explicitly provided, the variable is assumed to have no vocabulary.
var_name_to_prev_var_name: [Optional] Dict of variable names (strings) to name of the previously-trained variable in
ckpt_to_initialize_from. If not explicitly provided, the name of the variable is assumed to be same between previous checkpoint and current model.
ValueError: If the WarmStartSettings contains prev_var_name or VocabInfo configuration for variable names that are not used. This is to ensure a stronger check for variable configuration than relying on users to examine the logs.