def cv_type(self, value):
if isinstance(value, str):
try:
self._cv_type = sk_cv.__getattribute__(value)
except AttributeError:
raise AttributeError(f"'{value}' not in `sklearn.model_selection._split`")
else: # Assumed to be a valid CV class
self._cv_type = value
def validate_parameters(self):
"""Ensure the provided parameters are valid and properly formatted"""
#################### root_results_path ####################
if self.root_results_path is None:
G.warn('Received root_results_path=None. Results will not be stored at all.')
elif isinstance(self.root_results_path, str):
if not self.root_results_path.endswith(ASSETS_DIRNAME):
self.root_results_path = os.path.join(self.root_results_path, ASSETS_DIRNAME)
self.result_paths['root'] = self.root_results_path
if not os.path.exists(self.root_results_path):
os.makedirs(self.root_results_path, exist_ok=True)
else:
raise TypeError('root_results_path must be None or str, not {}: {}'.format(*type_val(self.root_results_path)))
#################### verbose ####################
if not isinstance(self.verbose, bool):
raise TypeError('verbose must be a boolean. Received {}: {}'.format(*type_val(self.verbose)))
#################### file_blacklist ####################
self.file_blacklist = validate_file_blacklist(self.file_blacklist)
#################### Train/Test Datasets ####################
if isinstance(self.train_dataset, str):
self.train_dataset = pd.read_csv(self.train_dataset)
if isinstance(self.test_dataset, str):
self.test_dataset = pd.read_csv(self.test_dataset)
#################### metrics_params/metrics_map ####################
if (self.metrics_map is not None) and ('metrics_map' in self.metrics_params.keys()):
raise ValueError(
'`metrics_map` may be provided as a kwarg, or as a key in `metrics_params`, but NOT BOTH. Received: ' +
F'\n `metrics_map`={self.metrics_map}\n `metrics_params`={self.metrics_params}'
)
else:
if self.metrics_map is None:
self.metrics_map = self.metrics_params['metrics_map']
self.metrics_params = {**dict(metrics_map=self.metrics_map), **self.metrics_params}
#################### cross_validation_type ####################
if isinstance(self.cross_validation_type, str):
try:
self.cross_validation_type = sk_cv.__getattribute__(self.cross_validation_type)
except AttributeError:
raise AttributeError('`sklearn.model_selection._split` has no attribute "{}".'.format(self.cross_validation_type))
#################### to_csv_params ####################
self.to_csv_params = {_k: _v for _k, _v in self.to_csv_params.items() if _k != 'path_or_buf'}
#################### cross_experiment_params ####################
self.cross_experiment_params = dict(
cross_validation_type=self.cross_validation_type,
runs=self.runs,
global_random_seed=self.global_random_seed,
random_seeds=self.random_seeds,
random_seed_bounds=self.random_seed_bounds,
)
#################### experiment_callbacks ####################
if not isinstance(self.experiment_callbacks, list):
self.experiment_callbacks = [self.experiment_callbacks]
for callback in self.experiment_callbacks:
if not isclass(callback):
raise TypeError(F'experiment_callbacks must be classes. Received {type(callback)}: {callback}')
if callback.__name__ != 'LambdaCallback':
raise ValueError(F'experiment_callbacks must be LambdaCallback instances, not {callback.__name__}: {callback}')
def validate_parameters(self):
"""Ensure the provided parameters are valid and properly formatted"""
#################### root_results_path ####################
if self.root_results_path is None:
G.warn(
"Received root_results_path=None. Results will not be stored at all."
)
elif isinstance(self.root_results_path, str):
if not self.root_results_path.endswith(ASSETS_DIRNAME):
self.root_results_path = os.path.join(self.root_results_path,
ASSETS_DIRNAME)
self.result_paths["root"] = self.root_results_path
if not os.path.exists(self.root_results_path):
make_dirs(self.root_results_path, exist_ok=True)
else:
raise TypeError(
f"root_results_path must be None or str, not {self.root_results_path}"
)
#################### target_column ####################
if isinstance(self.target_column, str):
self.target_column = [self.target_column]
#################### file_blacklist ####################
self.file_blacklist = validate_file_blacklist(self.file_blacklist)
if self.root_results_path is None:
self.file_blacklist = "ALL"
#################### Train/Test Datasets ####################
if isinstance(self.train_dataset, str):
self.train_dataset = pd.read_csv(self.train_dataset)
if isinstance(self.test_dataset, str):
self.test_dataset = pd.read_csv(self.test_dataset)
#################### metrics_params/metrics_map ####################
if (self.metrics_map is not None) and ("metrics_map"
in self.metrics_params.keys()):
raise ValueError(
"`metrics_map` may be provided as a kwarg, or as a `metrics_params` key, but NOT BOTH. Received: "
+
f"\n `metrics_map`={self.metrics_map}\n `metrics_params`={self.metrics_params}"
)
else:
if self.metrics_map is None:
self.metrics_map = self.metrics_params["metrics_map"]
self.metrics_map = format_metrics_map(self.metrics_map)
self.metrics_params = {
**dict(metrics_map=self.metrics_map),
**self.metrics_params
}
#################### cross_validation_type ####################
if isinstance(self.cross_validation_type, str):
try:
self.cross_validation_type = sk_cv.__getattribute__(
self.cross_validation_type)
except AttributeError:
raise AttributeError(
f"'{self.cross_validation_type}' not in `sklearn.model_selection._split`"
)
#################### to_csv_params ####################
self.to_csv_params = {
k: v
for k, v in self.to_csv_params.items() if k != "path_or_buf"
}
#################### cross_experiment_params ####################
self.cross_experiment_params = dict(
cross_validation_type=self.cross_validation_type,
runs=self.runs,
global_random_seed=self.global_random_seed,
random_seeds=self.random_seeds,
random_seed_bounds=self.random_seed_bounds,
)
#################### experiment_callbacks ####################
if not isinstance(self.experiment_callbacks, list):
self.experiment_callbacks = [self.experiment_callbacks]
for cb in self.experiment_callbacks:
if not isclass(cb):
raise TypeError(
f"experiment_callbacks must be classes, not {type(cb)}: {cb}"
)
if cb.__name__ != "LambdaCallback":
raise ValueError(
f"experiment_callbacks must be LambdaCallback instances, not {cb}"
)