def inspect(self, filename=None) -> Dict[str, str]:
"""
Get all details of the pipeline in human-readable format.
For a shorter human-readable representation, use MatPipe.summarize().
Args:
filename (str): An optional '.txt', '.yaml', '.yml', or '.json'
filename to use for saving the pipeline inspect.
Returns:
(dict): A dict of strings in human readable format. Contains a
string representation of every object and user parameter.
"""
attrs = return_attrs_recursively(self)
if filename:
save_dict_to_file(attrs, filename)
return attrs
def summarize(self, filename=None) -> Dict[str, str]:
"""
Get an executive summary of the most important parts of the pipeline.
Useful for understanding the pipeline at a high level.
For a more detailed human-readable representation, use MatPipe.inspect.
Args:
filename (str): An optional '.txt', '.yaml', '.yml', or '.json'
filename to use for saving the pipeline summarize.
Returns:
(dict): A dict of strings in human readable format. Contains a
string representation of every object and user parameter.
"""
cleaner_attrs = [
"encoder",
"feature_na_method",
"na_method_fit",
"na_method_transform",
"drop_na_targets",
]
cleaner_data = {
attr: str(getattr(self.cleaner, attr))
for attr in cleaner_attrs
}
reducer_attrs = ["reducers", "reducer_params"]
reducer_data = {
attr: str(getattr(self.reducer, attr))
for attr in reducer_attrs
}
attrs = {
"featurizers": self.autofeaturizer.featurizers,
"ml_model": str(self.learner.best_pipeline),
"feature_reduction": reducer_data,
"data_cleaning": cleaner_data,
"features": self.learner.features,
}
if filename:
save_dict_to_file(attrs, filename)
return attrs
def test_save_dict_to_file(self):
test_dict = {"a": "A", "b": 1, "c": [1, "q"], "d": {"m": [3, 4]}}
for ext in AMM_SUPPORTED_EXTS:
filename = self._get_remnant_path(ext)
save_dict_to_file(test_dict, filename=filename)
self.assertTrue(os.path.isfile(filename))