def add_complex_type_lookup_entry(self, path, key, value, hashed_value):
"""Add lookup entry in `key_attribute_lookup_dir` for a complex-typed parameter, linking the parameter `key`, its
`value`, and its `hashed_value`
Parameters
----------
path: Tuple
The path of keys that leads to `key`
key: Str
The parameter name
value: *
The value of the parameter `key`
hashed_value: Str
The hash produced for `value`"""
# TODO: Combine `path` and `key` to produce actual filepaths
shelve_params = ['model_initializer', 'cross_validation_type']
if isclass(value) or (key in shelve_params):
with shelve.open(os.path.join(self.key_attribute_lookup_dir, F'{key}'), flag='c') as shelf:
# NOTE: When reading from shelve file, DO NOT add the ".db" file extension
shelf[hashed_value] = value
elif isinstance(value, pd.DataFrame):
os.makedirs(os.path.join(self.key_attribute_lookup_dir, key), exist_ok=True)
value.to_csv(os.path.join(self.key_attribute_lookup_dir, key, F'{hashed_value}.csv'), index=False)
else: # Possible types: partial, function, *other
add_to_json(
file_path=os.path.join(self.key_attribute_lookup_dir, F'{key}.json'),
data_to_add=getsource(value), key=hashed_value, condition=lambda _: hashed_value not in _.keys(), default={},
)
def add_complex_type_lookup_entry(self, path, key, value, hashed_value):
"""Add lookup entry in `lookup_dir` for a complex-typed parameter, linking
the parameter `key`, its `value`, and its `hashed_value`
Parameters
----------
path: Tuple
The path of keys that leads to `key`
key: Str
The parameter name
value: *
The value of the parameter `key`
hashed_value: Str
The hash produced for `value`"""
shelve_params = ["model_initializer", "cv_type"]
lookup_path = partial(os.path.join, self.lookup_dir,
*[f"{_}" for _ in path])
if isclass(value) or (key in shelve_params):
make_dirs(lookup_path(), exist_ok=True)
with shelve.open(lookup_path(f"{key}"), flag="c") as s:
# NOTE: When reading from shelve file, DO NOT add the ".db" file extension
s[hashed_value] = value
elif isinstance(value, pd.DataFrame):
make_dirs(lookup_path(key), exist_ok=True)
value.to_csv(lookup_path(key, f"{hashed_value}.csv"), index=False)
else: # Possible types: partial, function, *other
add_to_json(
file_path=lookup_path(f"{key}.json"),
data_to_add=getsource(value),
key=hashed_value,
condition=lambda _: hashed_value not in _.keys(),
default={},
)
def save_result(self):
"""Save the cross-experiment, and hyperparameter keys, and update their tested keys entries"""
self.cross_experiment_key.save_key()
self.hyperparameter_key.save_key()
add_to_json(
file_path=
F'{self.hyperparameter_key.tested_keys_dir}/{self.cross_experiment_key.key}.json',
data_to_add=self.experiment_id,
key=self.hyperparameter_key.key,
condition=lambda _: self.hyperparameter_key.key in _.keys(),
append_value=True)
def save_key(self):
"""Create an entry in the dict contained in the file at :attr:`cross_experiment_key.key`,
whose key is :attr:`key`, and whose value is an empty list if :attr:`exists` is False"""
if not self.exists:
if self.cross_experiment_key.exists is False:
_err = "Cannot save hyperparameter_key: '{}', before cross_experiment_key '{}'"
raise ValueError(_err.format(self.key, self.cross_experiment_key.key))
key_path = f"{self.tested_keys_dir}/{self.cross_experiment_key.key}.json"
add_to_json(key_path, [], key=self.key, condition=lambda _: self.key not in _.keys())
self.exists = True
G.log(f'Saved {self.key_type}_key: "{self.key}"', 4)
else:
G.log(f'{self.key_type}_key "{self.key}" already exists - Skipped saving', 4)
def save_key(self):
"""Create an entry in the dict corresponding to the file at :attr:`cross_experiment_key.key`, whose key is :attr:`key`,
and whose value is an empty list if :attr:`exists` is False"""
if not self.exists:
if self.cross_experiment_key.exists is False:
raise ValueError('Cannot save hyperparameter_key: "{}", before cross_experiment_key "{}" has been saved'.format(
self.key, self.cross_experiment_key.key
))
key_path = F'{self.tested_keys_dir}/{self.cross_experiment_key.key}.json'
add_to_json(file_path=key_path, data_to_add=[], key=self.key, condition=lambda _: self.key not in _.keys())
self.exists = True
G.log(F'Saved {self.key_type}_key: "{self.key}"')
else:
G.log(F'{self.key_type}_key "{self.key}" already exists - Skipped saving')
def add_complex_type_lookup_entry(self, path, key, value, hashed_value):
"""Add lookup entry in `lookup_dir` for a complex-typed parameter, linking
the parameter `key`, its `value`, and its `hashed_value`
Parameters
----------
path: Tuple
The path of keys that leads to `key`
key: Str
The parameter name
value: *
The value of the parameter `key`
hashed_value: Str
The hash produced for `value`"""
shelve_params = ["model_initializer", "cv_type"]
lookup_path = partial(os.path.join, self.lookup_dir,
*[f"{_}" for _ in path])
if isclass(value) or (key in shelve_params):
make_dirs(lookup_path(), exist_ok=True)
with shelve.open(lookup_path(f"{key}"), flag="c") as s:
# NOTE: When reading from shelve file, DO NOT add the ".db" file extension
try:
s[hashed_value] = value
except PicklingError:
# "is not the same object" error can be raised due to `Mirror`/`TranslateTrace`
# Instead of saving the object that raised the error, save `getsourcelines`
# Source lines of traced object are identical to those of its un-traced original
s[hashed_value] = getsourcelines(value)
except Exception:
raise
elif isinstance(value, pd.DataFrame):
make_dirs(lookup_path(key), exist_ok=True)
value.to_csv(lookup_path(key, f"{hashed_value}.csv"), index=False)
else: # Possible types: partial, function, *other
add_to_json(
file_path=lookup_path(f"{key}.json"),
data_to_add=getsource(value),
key=hashed_value,
condition=lambda _: hashed_value not in _.keys(),
default={},
)
