def create_trial_if_possible(self, experiment_spec: Dict,
output_path: str) -> Optional[Trial]:
logger.debug("creating trial")
trial_id = Trial.generate_id()
suggested_config = self.searcher.suggest(trial_id)
if suggested_config == Searcher.FINISHED:
self._finished = True
logger.debug("Searcher has finished.")
return
if suggested_config is None:
return
spec = copy.deepcopy(experiment_spec)
spec["config"] = merge_dicts(spec["config"],
copy.deepcopy(suggested_config))
# Create a new trial_id if duplicate trial is created
flattened_config = resolve_nested_dict(spec["config"])
self._counter += 1
tag = "{0}_{1}".format(str(self._counter),
format_vars(flattened_config))
trial = create_trial_from_spec(
spec,
output_path,
self._parser,
evaluated_params=flatten_dict(suggested_config),
experiment_tag=tag,
trial_id=trial_id,
)
return trial
def _read_from_disk(self, policy_id):
"""Reads a policy ID from disk and re-adds it to the cache.
"""
# Make sure this policy ID is not in the cache right now.
assert policy_id not in self.cache
# Read policy state from disk.
with open(self.path + "/" + policy_id + self.extension, "rb") as f:
policy_state = pickle.load(f)
# Get class and config override.
merged_conf = merge_dicts(self.policy_config,
self.policy_specs[policy_id].config)
# Create policy object (from its spec: cls, obs-space, act-space,
# config).
self.create_policy(
policy_id,
self.policy_specs[policy_id].policy_class,
self.policy_specs[policy_id].observation_space,
self.policy_specs[policy_id].action_space,
self.policy_specs[policy_id].config,
merged_conf,
)
# Restore policy's state.
policy = self[policy_id]
policy.set_state(policy_state)
def _train(self):
"""Calls `self.trainer.train()` and `self.trainer.validate()` once.
You may want to override this if using a custom LR scheduler.
"""
train_stats = self.trainer.train(max_retries=10, profile=True)
validation_stats = self.trainer.validate(profile=True)
stats = merge_dicts(train_stats, validation_stats)
return stats
def step(self):
"""Calls `self.trainer.train()` and `self.trainer.validate()` once."""
if self._implements_method("_train"):
raise DeprecationWarning(
"Trainable._train is deprecated and is now removed."
"Override Trainable.step instead.")
train_stats = self.trainer.train(max_retries=0, profile=True)
validation_stats = self.trainer.validate(profile=True)
stats = merge_dicts(train_stats, validation_stats)
return stats
def step(self):
"""Calls `self.trainer.train()` and `self.trainer.validate()` once."""
if self._is_overridden("_train"):
raise DeprecationWarning(
"Trainable._train is deprecated and will be "
"removed in "
"a future version of Ray. Override Trainable.step instead.")
train_stats = self.trainer.train(max_retries=10, profile=True)
validation_stats = self.trainer.validate(profile=True)
stats = merge_dicts(train_stats, validation_stats)
return stats
def _train(self):
"""Calls `self.trainer.train()` and `self.trainer.validate()` once.
You may want to override this if using a custom LR scheduler.
"""
train_stats = self.trainer.train(max_retries=10, profile=True)
validation_stats = self.trainer.validate(profile=True)
stats = merge_dicts(train_stats, validation_stats)
stats.update(hostplan=self.trainer.config.get(
"all_fluid_trial_resources", {}), )
return stats
def _train(self):
train_stats = self.trainer.train(max_retries=0, profile=True)
validation_stats = self.trainer.validate(profile=True)
stats = merge_dicts(train_stats, validation_stats)
return stats
