def test_integration(self):
res = self.extractor.extract_features(self.files)
self.assertIsNotNone(res, "Failed to parse files.")
X, y, _, = res
train_X, test_X, train_y, test_y = \
model_selection.train_test_split(X, y, random_state=1989)
model = tree.DecisionTreeClassifier(min_samples_leaf=26,
random_state=1989,
max_depth=None,
max_features="auto",
min_samples_split=2)
model.fit(train_X, train_y)
rules = TrainableRules(
base_model_name="sklearn.tree.DecisionTreeClassifier",
prune_branches_algorithms=[],
prune_attributes=False,
min_samples_leaf=26,
random_state=1989,
max_depth=None,
max_features="auto",
min_samples_split=2,
confidence_threshold=0)
rules.fit(train_X, train_y)
model_score_train = model.score(train_X, train_y)
model_score_test = model.score(test_X, test_y)
rules_score_train = rules.score(train_X, train_y)
rules_score_test = rules.score(test_X, test_y)
self.assertEqual(rules_score_train, model_score_train)
self.assertEqual(rules_score_test, model_score_test)
def test_integration(self):
X, y, _ = self.extractor.extract_features(self.files)
train_X, test_X, train_y, test_y = \
model_selection.train_test_split(X, y, random_state=1989)
model = tree.DecisionTreeClassifier(min_samples_leaf=26,
random_state=1989,
max_depth=None,
max_features="auto",
min_samples_split=2)
model.fit(train_X, train_y)
rules = TrainableRules("sklearn.tree.DecisionTreeClassifier",
prune_branches_algorithms=[],
prune_attributes=False,
min_samples_leaf=26,
random_state=1989,
max_depth=None,
max_features="auto",
min_samples_split=2)
rules.fit(train_X, train_y)
model_score_train = model.score(train_X, train_y)
model_score_test = model.score(test_X, test_y)
rules_score_train = rules.score(train_X, train_y)
rules_score_test = rules.score(test_X, test_y)
self.assertEqual(rules_score_train, model_score_train)
self.assertEqual(rules_score_test, model_score_test)
def test_budget(budget):
rules = TrainableRules(
"sklearn.tree.DecisionTreeClassifier",
prune_branches_algorithms=["top-down-greedy"],
prune_attributes=False,
top_down_greedy_budget=(False, budget),
random_state=1989)
rules.fit(self.train_x, self.train_y)
return rules.score(self.train_x, self.train_y)
def test_predict_winner_indices(self):
rules = TrainableRules("sklearn.tree.DecisionTreeClassifier",
prune_branches_algorithms=[],
prune_attributes=False,
min_samples_leaf=26,
random_state=1989)
rules.fit(self.train_x, self.train_y)
pred_y, winners = rules.rules.predict(self.train_x,
return_winner_indices=True)
for ycls, w in zip(pred_y, winners):
self.assertEqual(ycls, rules.rules.rules[w].stats.cls)
def test_tree_attr_pruning(self):
model = tree.DecisionTreeClassifier(min_samples_leaf=26,
random_state=1989)
model = model.fit(self.train_x, self.train_y)
rules = TrainableRules("sklearn.tree.DecisionTreeClassifier",
prune_branches_algorithms=[],
prune_attributes=True,
min_samples_leaf=26,
random_state=1989)
rules.fit(self.train_x, self.train_y)
tree_score = model.score(self.test_x, self.test_y)
rules_score = rules.score(self.test_x, self.test_y)
self.assertGreater(rules_score * 1.1, tree_score)
def test_tree_no_pruning(self):
model = tree.DecisionTreeClassifier(min_samples_leaf=26,
random_state=1989)
model = model.fit(self.train_x, self.train_y)
rules = TrainableRules(
base_model_name="sklearn.tree.DecisionTreeClassifier",
prune_branches_algorithms=[],
confidence_threshold=0,
prune_attributes=False,
min_samples_leaf=26,
random_state=1989)
rules.fit(self.train_x, self.train_y)
tree_score = model.score(self.train_x, self.train_y)
rules_score = rules.score(self.train_x, self.train_y)
self.assertGreater(rules_score * 1.1, tree_score)
def test_forest_no_pruning(self):
model = ensemble.RandomForestClassifier(n_estimators=50,
min_samples_leaf=26,
random_state=1989)
model = model.fit(self.train_x, self.train_y)
rules = TrainableRules("sklearn.ensemble.RandomForestClassifier",
prune_branches_algorithms=[],
prune_attributes=False,
n_estimators=50,
min_samples_leaf=26,
random_state=1989)
rules.fit(self.train_x, self.train_y)
forest_score = model.score(self.train_x, self.train_y)
rules_score = rules.score(self.train_x, self.train_y)
self.assertGreater(rules_score * 1.1, forest_score)
def setUp(self):
(self.train_x, self.test_x, self.train_y,
self.test_y), _, _ = load_abalone_data()
self.config = {
"trainable_rules": {
"base_model_name": "sklearn.tree.DecisionTreeClassifier",
"prune_branches_algorithms": [],
"prune_attributes": False,
"min_samples_leaf": 26,
"random_state": 1989,
},
}
trainer = TrainableRules(**self.config["trainable_rules"],
origin_config=self.config)
trainer.fit(self.test_x, self.test_y)
self.rules = trainer.rules
self.fm = FormatModel().load(
os.path.join(os.path.dirname(__file__), "model_jquery.asdf"))
self.maxDiff = None
def train(cls, ptr: ReferencePointer, config: Dict[str, Any],
data_request_stub: DataStub, **data) -> AnalyzerModel:
"""
Train a model given the files available.
:param ptr: Git repository state pointer.
:param config: configuration dict.
:param data: contains "files" - the list of files in the pointed state.
:param data_request_stub: connection to the Lookout data retrieval service, not used.
:return: AnalyzerModel containing the learned rules, per language.
"""
config = cls._load_train_config(config)
cls.log.info("train %s %s %s", ptr.url, ptr.commit,
pformat(config, width=4096, compact=True))
files_by_language = cls._files_by_language(data["files"])
model = FormatModel().construct(cls, ptr)
for language, files in files_by_language.items():
language = language.lower()
try:
fe = FeatureExtractor(
language=language,
siblings_window=config["siblings_window"],
parents_depth=config["parents_depth"])
except ImportError:
cls.log.warning("skipped %d %s files - not supported",
len(files), language)
continue
else:
cls.log.info("training on %d %s files", len(files), language)
# we sort to make the features reproducible
X, y, _ = fe.extract_features(f[1] for f in sorted(files.items()))
lower_bound_instances = config["lower_bound_instances"]
if X.shape[0] < lower_bound_instances:
cls.log.warning("skipped %d %s files: too few samples (%d/%d)",
len(files), language, X.shape[0],
lower_bound_instances)
continue
cls.log.debug("training the rules model")
bscv = BayesSearchCV(TrainableRules(
prune_branches_algorithms=config["prune_branches_algorithms"],
prune_attributes=config["prune_attributes"],
top_down_greedy_budget=config["top_down_greedy_budget"],
uncertain_attributes=config["uncertain_attributes"],
prune_dataset_ratio=config["prune_dataset_ratio"],
n_estimators=config["n_estimators"],
random_state=config["random_state"]), {
"base_model_name":
Categorical([
"sklearn.ensemble.RandomForestClassifier",
"sklearn.tree.DecisionTreeClassifier"
]),
"max_depth":
Categorical([None, 5, 10]),
"max_features":
Categorical([None, "auto"]),
"min_samples_split":
Integer(2, 20),
"min_samples_leaf":
Integer(1, 20)
},
n_jobs=-1,
n_iter=config["n_iter"],
random_state=config["random_state"])
bscv.fit(X, y)
cls.log.debug("score of the best estimator found: %.3f",
bscv.best_score_)
cls.log.debug("params of the best estimator found: %s",
str(bscv.best_params_))
cls.log.debug("training the model with complete data")
trainable_rules = TrainableRules(
prune_branches_algorithms=["reduced-error"],
prune_attributes=True,
random_state=42,
uncertain_attributes=True,
**bscv.best_params_)
trainable_rules.fit(X, y)
model[language] = trainable_rules.rules
cls.log.info("trained %s", model)
return model
def train(cls, ptr: ReferencePointer, config: Mapping[str, Any], data_service: DataService,
files: Iterator[File], **data) -> FormatModel:
"""
Train a model given the files available.
:param ptr: Git repository state pointer.
:param config: configuration dict.
:param data: contains "files" - the list of files in the pointed state.
:param data_service: connection to the Lookout data retrieval service.
:param files: iterator of File records from the data service.
:return: AnalyzerModel containing the learned rules, per language.
"""
_log = logging.getLogger(cls.__name__)
train_config = cls._load_config(config)["train"]
_log.info("train %s %s %s %s", __version__, ptr.url, ptr.commit,
pformat(train_config, width=4096, compact=True))
model = FormatModel().generate(cls, ptr)
for language, files in files_by_language(files).items():
try:
lang_config = train_config[language]
except KeyError:
_log.warning("language %s is not supported, skipped", language)
continue
_log.info("effective train config for %s:\n%s", language,
pformat(lang_config, width=120, compact=True))
random_state = lang_config["random_state"]
files = filter_files(
files, lang_config["line_length_limit"], lang_config["overall_size_limit"],
random_state, _log)
submit_event("%s.train.%s.files" % (cls.name, language), len(files))
if len(files) == 0:
_log.info("zero files after filtering, language %s is skipped.", language)
continue
try:
fe = FeatureExtractor(language=language, **lang_config["feature_extractor"])
except ImportError:
_log.warning("skipped %d %s files - not supported", len(files), language)
continue
else:
_log.info("training on %d %s files", len(files), language)
train_files, test_files = FormatAnalyzer.split_train_test(
files, lang_config["test_dataset_ratio"], random_state=random_state)
# ensure that the features are reproducible
train_files = sorted(train_files, key=lambda x: x.path)
test_files = sorted(test_files, key=lambda x: x.path)
X_train, y_train, _ = fe.extract_features(train_files)
X_train, selected_features = fe.select_features(X_train, y_train)
if test_files:
X_test, y_test, _ = fe.extract_features(test_files)
if lang_config["test_dataset_ratio"]:
_log.debug("Real test ratio is %.3f",
X_test.shape[0] / (X_test.shape[0] + X_train.shape[0])
if test_files else 0)
lang_config["feature_extractor"]["selected_features"] = selected_features
lang_config["feature_extractor"]["label_composites"] = fe.labels_to_class_sequences
lower_bound_instances = lang_config["lower_bound_instances"]
if X_train.shape[0] < lower_bound_instances:
_log.warning("skipped %d %s files: too few samples (%d/%d)",
len(files), language, X_train.shape[0], lower_bound_instances)
continue
_log.info("extracted %d samples to train, searching for the best hyperparameters",
X_train.shape[0])
optimizer = Optimizer(**lang_config["optimizer"], random_state=random_state)
best_score, best_params = optimizer.optimize(X_train, y_train)
if _log.isEnabledFor(logging.DEBUG):
_log.debug("score of the best estimator found: %.6f", best_score)
_log.debug("params of the best estimator found: %s", str(best_params))
_log.debug("training the model with complete data")
else:
_log.info("finished hyperopt at %.6f, training the full model", -best_score)
lang_config["trainable_rules"].update(best_params)
trainable_rules = TrainableRules(**lang_config["trainable_rules"],
random_state=random_state,
origin_config=lang_config)
trainable_rules.fit(X_train, y_train)
importances = trainable_rules.feature_importances_
_log.debug(
"feature importances from %s:\n\t%s",
lang_config["trainable_rules"]["base_model_name"],
"\n\t".join("%-55s %.5E" % (fe.feature_names[i], importances[i])
for i in numpy.argsort(-importances)[:25] if importances[i] > 1e-5))
trainable_rules.prune_categorical_attributes(fe)
_log.info("obtained %d rules, generating the classification report",
len(trainable_rules.rules))
trainable_rules.rules.generate_classification_report(
X_train, y_train, "train", fe.composite_class_representations)
if test_files:
trainable_rules.rules.generate_classification_report(
X_test, y_test, "test", fe.composite_class_representations)
submit_event("%s.train.%s.rules" % (cls.name, language), len(trainable_rules.rules))
if trainable_rules.rules.rules:
model[language] = trainable_rules.rules
else:
_log.warning("model for %s has 0 rules. Skipped.", language)
_log.info("trained %s", model)
return model
def train(cls, ptr: ReferencePointer, config: Mapping[str, Any],
data_service: DataService, **data) -> FormatModel:
"""
Train a model given the files available.
:param ptr: Git repository state pointer.
:param config: configuration dict.
:param data: contains "files" - the list of files in the pointed state.
:param data_service: connection to the Lookout data retrieval service.
:return: AnalyzerModel containing the learned rules, per language.
"""
_log = logging.getLogger(cls.__name__)
_log.info("train %s %s %s", ptr.url, ptr.commit,
pformat(config, width=4096, compact=True))
model = FormatModel().construct(cls, ptr)
config = cls._load_train_config(config)
for language, files in files_by_language(data["files"]).items():
try:
lang_config = config[language]
except KeyError:
_log.warning("language %s is not supported, skipped", language)
continue
files = filter_files(files, lang_config["line_length_limit"], _log)
submit_event("%s.train.%s.files" % (cls.name, language),
len(files))
if len(files) == 0:
_log.info(
"zero files after filtering, language %s is skipped.",
language)
continue
try:
fe = FeatureExtractor(language=language,
**lang_config["feature_extractor"])
except ImportError:
_log.warning("skipped %d %s files - not supported", len(files),
language)
continue
else:
_log.info("training on %d %s files", len(files), language)
# we sort to make the features reproducible
X, y, _ = fe.extract_features(sorted(files, key=lambda x: x.path))
X, selected_features = fe.select_features(X, y)
lang_config["feature_extractor"][
"selected_features"] = selected_features
lang_config["feature_extractor"][
"label_composites"] = fe.labels_to_class_sequences
lower_bound_instances = lang_config["lower_bound_instances"]
if X.shape[0] < lower_bound_instances:
_log.warning("skipped %d %s files: too few samples (%d/%d)",
len(files), language, X.shape[0],
lower_bound_instances)
continue
_log.debug("training the rules model")
optimizer = Optimizer(
n_jobs=lang_config["n_jobs"],
n_iter=lang_config["n_iter"],
cv=lang_config["cv"],
random_state=lang_config["trainable_rules"]["random_state"])
best_score, best_params = optimizer.optimize(X, y)
_log.debug("score of the best estimator found: %.6f", best_score)
_log.debug("params of the best estimator found: %s",
str(best_params))
_log.debug("training the model with complete data")
lang_config["trainable_rules"].update(best_params)
trainable_rules = TrainableRules(**lang_config["trainable_rules"],
origin_config=lang_config)
trainable_rules.fit(X, y)
importances = trainable_rules.feature_importances_
_log.debug(
"feature importances from %s:\n\t%s",
lang_config["trainable_rules"]["base_model_name"], "\n\t".join(
"%-55s %.5E" % (fe.feature_names[i], importances[i])
for i in numpy.argsort(-importances)[:25]
if importances[i] > 1e-5))
submit_event("%s.train.%s.rules" % (cls.name, language),
len(trainable_rules.rules))
# TODO(vmarkovtsev): save the achieved precision, recall, etc. to the model
# throw away imprecise classes
if trainable_rules.rules.rules:
model[language] = trainable_rules.rules
else:
_log.warning("model for %s has 0 rules. Skipping.", language)
_log.info("trained %s", model)
return model