def test_flatten(self):
flat_config_str = 'dataset.batch_size=32,dataset.epoch_size=96,dataset.fn=TokenCopyTask,dataset.min_length=10,dataset.vocab_size=10,model.cell.fn=DAGCell,model.optimizer.learning_rate=0.001' # pylint: disable=line-too-long
flat_config = {
'dataset.min_length': 10,
'dataset.fn': 'TokenCopyTask',
'model.cell.fn': 'DAGCell',
'model.optimizer.learning_rate': 0.001,
'dataset.vocab_size': 10,
'dataset.epoch_size': 96,
'dataset.batch_size': 32
}
nested_config = {
'dataset': {
'batch_size': 32,
'epoch_size': 96,
'fn': 'TokenCopyTask',
'min_length': 10,
'vocab_size': 10
},
'model': {
'optimizer': {
'learning_rate': 0.001
},
'cell': {
'fn': 'DAGCell'
}
}
}
self.assertAllEqual(flat_config,
configurable.flatten_config(nested_config))
self.assertAllEqual(flat_config_str,
configurable.config_to_string(nested_config))
self.assertAllEqual(configurable.unflatten_dict(flat_config),
nested_config)
def _create_config(hparams):
"""Create trial config and save to disk.
Args:
hparams: Nondefault params to merge
Returns:
A configurable object `spec` instantiated from the final config.
spec.config will return the config.
"""
hparams = hparams or contrib_training.HParams()
tuned_config = configurable.unflatten_dict(hparams.values())
pprinter = pprint.PrettyPrinter()
tf.logging.info('Provided extra params:\n%s',
pprinter.pformat(tuned_config))
try:
merged_config = configurable.merge(default_config, tuned_config)
tf.logging.info('Tuned default config:\n%s', merged_config)
except TypeError as e:
tf.logging.info(
'Do not provide same config in both config string and vizier.'
' This may lead to type errors.')
raise e
# Instantiate a ConfigurableExperiment object.
experiment_spec = configurable.Configurable.initialize(merged_config)
tf.logging.info('Final config:\n%s', experiment_spec.config)
return experiment_spec
