def extract_roots(self, model, rev_ids, caches=None):
"""
Extracts a mapping of root datasources capabile of generating the
features needed for a particular model.
:Parameters:
rev_ids : int | `iterable`
Revision IDs to extract for
model : str
The name of a
:class:`~revscoring.scorer_models.scorer_model.ScorerModel` to
extract the roots for
"""
features = self[model].features
root_ds = [d for d in dependencies.dig(features)
if isinstance(d, Datasource)]
error_root_vals = self.extractor.extract(rev_ids, root_ds,
caches=caches)
root_ds_caches = {}
for rev_id, (error, root_vals) in zip(rev_ids, error_root_vals):
if error is None:
root_ds_caches[rev_id] = \
(None, {rd: rv for rd, rv in zip(root_ds, root_vals)})
else:
root_ds_caches[rev_id] = (error, None)
return root_ds_caches
def extract_roots(self, model, rev_ids, caches=None):
"""
Extracts a mapping of root datasources capabile of generating the
features needed for a particular model.
:Parameters:
rev_ids : int | `iterable`
Revision IDs to extract for
model : str
The name of a :class:`~revscoring.ScorerModel` to
extract the roots for
"""
features = self[model].features
root_ds = [
d for d in dependencies.dig(features) if isinstance(d, Datasource)
]
error_root_vals = self.extractor.extract(rev_ids,
root_ds,
caches=caches)
root_ds_caches = {}
for rev_id, (error, root_vals) in zip(rev_ids, error_root_vals):
if error is None:
root_ds_map = {rd: rv for rd, rv in zip(root_ds, root_vals)}
if caches is not None:
root_ds_map.update(caches.get(rev_id, {}))
root_ds_caches[rev_id] = (None, root_ds_map)
else:
root_ds_caches[rev_id] = (error, None)
return root_ds_caches
def _generate_root_datasources(self, model_names):
for model_name in model_names:
for dependency in dependencies.dig(self.model_features(model_name)):
if isinstance(dependency, Datasource):
yield dependency
def load_model_and_queue(self, q, config, key):
model = Model.from_config(config, key)
# Just return the model info and the root of the features
q.put((model.info, list(dig(model.features))))
def _generate_root_datasources(self, model_names):
for model_name in model_names:
for dependency in dependencies.dig(self.model_features(model_name)):
if isinstance(dependency, Datasource):
yield dependency