def test_CommitExtractor():
CommitExtractor([reviewers_num(), author_experience()], [fileref(), url()])
with pytest.raises(AssertionError):
CommitExtractor([reviewers_num(), author_experience()],
[fileref(), fileref()])
with pytest.raises(AssertionError):
CommitExtractor(
[author_experience(), author_experience()],
[fileref(), url()])
def __init__(self, lemmatization=False):
BugCoupleModel.__init__(self, lemmatization)
self.calculate_importance = False
cleanup_functions = [
feature_cleanup.responses(),
feature_cleanup.hex(),
feature_cleanup.dll(),
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
feature_cleanup.crash(),
]
self.extraction_pipeline = Pipeline(
[
("bug_extractor", bug_features.BugExtractor([], cleanup_functions)),
(
"union",
ColumnTransformer([("text", self.text_vectorizer(), "text")]),
),
]
)
self.clf = LinearSVCWithLabelEncoding(LinearSVC())
def __init__(self, lemmatization=False):
IssueModel.__init__(self, lemmatization)
feature_extractors = [
issue_features.comment_count(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline([
(
"issue_extractor",
issue_features.IssueExtractor(feature_extractors,
cleanup_functions),
),
(
"union",
ColumnTransformer([
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.0001), "title"),
(
"first_comment",
self.text_vectorizer(min_df=0.0001),
"first_comment",
),
]),
),
])
self.clf = xgboost.XGBClassifier(n_jobs=utils.get_physical_cpu_count())
self.clf.set_params(predictor="cpu_predictor")
def __init__(self, lemmatization=False):
BugModel.__init__(self, lemmatization, commit_data=True)
self.cross_validation_enabled = False
self.sampler = RandomUnderSampler(random_state=0)
feature_extractors = [
bug_features.has_str(),
bug_features.has_regression_range(),
bug_features.severity(),
bug_features.keywords({"dev-doc-needed", "dev-doc-complete"}),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.product(),
bug_features.component(),
bug_features.commit_added(),
bug_features.commit_deleted(),
bug_features.commit_types(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"bug_extractor",
bug_features.BugExtractor(
feature_extractors,
cleanup_functions,
rollback=True,
rollback_when=self.rollback,
commit_data=True,
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(), "title"),
("comments", self.text_vectorizer(), "comments"),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=utils.get_physical_cpu_count())
self.clf.set_params(predictor="cpu_predictor")
def test_fileref():
tests = [(
"Some random filenames are file1.py , file2.cpp and file3.json",
"Some random filenames are __FILE_REFERENCE__ , __FILE_REFERENCE__ and __FILE_REFERENCE__",
)]
for orig_text, cleaned_text in tests:
assert feature_cleanup.fileref()(orig_text) == cleaned_text
def __init__(self, lemmatization=False, interpretable=False):
CommitModel.__init__(self, lemmatization)
self.required_dbs.append(BUG_INTRODUCING_COMMITS_DB)
self.store_dataset = True
self.sampler = RandomUnderSampler(random_state=0)
feature_extractors = [
commit_features.source_code_file_size(),
commit_features.other_file_size(),
commit_features.test_file_size(),
commit_features.source_code_added(),
commit_features.other_added(),
commit_features.test_added(),
commit_features.source_code_deleted(),
commit_features.other_deleted(),
commit_features.test_deleted(),
commit_features.author_experience(),
commit_features.reviewer_experience(),
commit_features.reviewers_num(),
commit_features.component_touched_prev(),
commit_features.directory_touched_prev(),
commit_features.file_touched_prev(),
commit_features.types(),
commit_features.files(),
commit_features.components(),
commit_features.components_modified_num(),
commit_features.directories(),
commit_features.directories_modified_num(),
commit_features.source_code_files_modified_num(),
commit_features.other_files_modified_num(),
commit_features.test_files_modified_num(),
commit_features.functions_touched_num(),
commit_features.functions_touched_size(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
column_transformers = [("data", DictVectorizer(), "data")]
if not interpretable:
column_transformers.append(
("desc", self.text_vectorizer(min_df=0.0001), "desc"))
self.extraction_pipeline = Pipeline([
(
"commit_extractor",
commit_features.CommitExtractor(feature_extractors,
cleanup_functions),
),
("union", ColumnTransformer(column_transformers)),
])
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
def __init__(self, lemmatization=False, historical=False):
BugModel.__init__(self, lemmatization)
self.calculate_importance = False
feature_extractors = [
bug_features.has_str(),
bug_features.severity(),
# Ignore keywords that would make the ML completely skewed
# (we are going to use them as 100% rules in the evaluation phase).
bug_features.keywords(set(KEYWORD_DICT.keys())),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.title(),
bug_features.blocked_bugs_number(),
bug_features.ever_affected(),
bug_features.affected_then_unaffected(),
bug_features.product(),
bug_features.component(),
]
cleanup_functions = [
feature_cleanup.url(),
feature_cleanup.fileref(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline([
(
"bug_extractor",
bug_features.BugExtractor(feature_extractors,
cleanup_functions),
),
(
"union",
ColumnTransformer([
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.001), "title"),
(
"first_comment",
self.text_vectorizer(min_df=0.001),
"first_comment",
),
(
"comments",
self.text_vectorizer(min_df=0.001),
"comments",
),
]),
),
])
self.clf = OneVsRestClassifier(xgboost.XGBClassifier(n_jobs=16))
def __init__(self, lemmatization=False):
BugModel.__init__(self, lemmatization)
self.sampler = RandomUnderSampler(random_state=0)
feature_extractors = [
bug_features.has_str(),
bug_features.has_regression_range(),
bug_features.severity(),
bug_features.keywords({"dev-doc-needed", "dev-doc-complete"}),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.title(),
bug_features.product(),
bug_features.component(),
bug_features.commit_added(),
bug_features.commit_deleted(),
bug_features.commit_types(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"bug_extractor",
bug_features.BugExtractor(
feature_extractors,
cleanup_functions,
rollback=True,
rollback_when=self.rollback,
commit_data=True,
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(), "title"),
("comments", self.text_vectorizer(), "comments"),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
def __init__(self, lemmatization=False):
BugModel.__init__(self, lemmatization)
self.sampler = BorderlineSMOTE(random_state=0)
self.calculate_importance = False
feature_extractors = [
bug_features.has_str(),
bug_features.has_regression_range(),
bug_features.severity(),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.product(),
# TODO: We would like to use the component at the time of filing too,
# but we can't because the rollback script doesn't support changes to
# components yet.
# bug_features.component(),
bug_features.num_words_title(),
bug_features.num_words_comments(),
bug_features.keywords(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"bug_extractor",
bug_features.BugExtractor(
feature_extractors, cleanup_functions, rollback=True
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.0001), "title"),
(
"comments",
self.text_vectorizer(min_df=0.0001),
"comments",
),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
def __init__(self, lemmatization=False):
BugModel.__init__(self, lemmatization)
self.cross_validation_enabled = False
self.calculate_importance = False
feature_extractors = [
bug_features.has_str(),
bug_features.severity(),
bug_features.keywords(),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.title(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"bug_extractor",
bug_features.BugExtractor(
feature_extractors, cleanup_functions, rollback=True
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.0001), "title"),
(
"comments",
self.text_vectorizer(min_df=0.0001),
"comments",
),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
self.CONFLATED_COMPONENTS_INVERSE_MAPPING = {
v: k for k, v in self.CONFLATED_COMPONENTS_MAPPING.items()
}
def __init__(self, lemmatization: bool = False) -> None:
CommitModel.__init__(self, lemmatization)
self.calculate_importance = False
self.training_dbs += [bugzilla.BUGS_DB]
self.sampler = RandomUnderSampler(random_state=0)
feature_extractors = [
commit_features.source_code_file_size(),
commit_features.other_file_size(),
commit_features.test_file_size(),
commit_features.source_code_added(),
commit_features.other_added(),
commit_features.test_added(),
commit_features.source_code_deleted(),
commit_features.other_deleted(),
commit_features.test_deleted(),
commit_features.reviewers_num(),
commit_features.types(),
commit_features.files(),
commit_features.components(),
commit_features.components_modified_num(),
commit_features.directories(),
commit_features.directories_modified_num(),
commit_features.source_code_files_modified_num(),
commit_features.other_files_modified_num(),
commit_features.test_files_modified_num(),
commit_features.functions_touched_num(),
commit_features.functions_touched_size(),
commit_features.source_code_file_metrics(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline([
(
"commit_extractor",
commit_features.CommitExtractor(feature_extractors,
cleanup_functions),
),
(
"union",
ColumnTransformer([
("data", DictVectorizer(), "data"),
("desc", self.text_vectorizer(min_df=0.0001), "desc"),
]),
),
])
self.clf = xgboost.XGBClassifier(n_jobs=utils.get_physical_cpu_count())
self.clf.set_params(predictor="cpu_predictor")
def __init__(self, lemmatization=False):
BugModel.__init__(self, lemmatization)
self.cross_validation_enabled = False
self.calculate_importance = False
feature_extractors = [
bug_features.has_str(),
bug_features.severity(),
bug_features.keywords(),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"bug_extractor",
bug_features.BugExtractor(
feature_extractors, cleanup_functions, rollback=True
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.0001), "title"),
(
"comments",
self.text_vectorizer(min_df=0.0001),
"comments",
),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
self.CONFLATED_COMPONENTS_INVERSE_MAPPING = {
v: k for k, v in self.CONFLATED_COMPONENTS_MAPPING.items()
}
def __init__(self, lemmatization=False):
CommitModel.__init__(self, lemmatization)
self.calculate_importance = False
self.sampler = RandomUnderSampler(random_state=0)
feature_extractors = [
commit_features.file_size(),
commit_features.test_added(),
commit_features.added(),
commit_features.deleted(),
commit_features.test_deleted(),
commit_features.author_experience(),
commit_features.reviewer_experience(),
commit_features.reviewers_num(),
commit_features.component_touched_prev(),
commit_features.directory_touched_prev(),
commit_features.file_touched_prev(),
commit_features.types(),
commit_features.components(),
commit_features.components_modified_num(),
commit_features.directories(),
commit_features.directories_modified_num(),
commit_features.files(),
commit_features.files_modified_num(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"commit_extractor",
commit_features.CommitExtractor(
feature_extractors, cleanup_functions
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("desc", self.text_vectorizer(), "desc"),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
def __init__(self,
training_size=14000,
lemmatization=False,
cleanup_urls=True):
self.num_duplicates = training_size // 2
self.num_nondups_nondups = self.num_dup_nondups = training_size // 4
BugCoupleModel.__init__(self, lemmatization)
self.calculate_importance = False
feature_extractors = [
bug_features.is_same_product(),
bug_features.is_same_component(),
bug_features.is_same_platform(),
bug_features.is_same_version(),
bug_features.is_same_os(),
bug_features.is_same_target_milestone(),
bug_features.is_first_affected_same(),
bug_features.couple_common_words_comments(),
bug_features.couple_delta_creation_date(),
bug_features.couple_common_keywords(),
bug_features.couple_common_whiteboard_keywords(),
bug_features.couple_common_words_summary(),
]
cleanup_functions = [
feature_cleanup.responses(),
feature_cleanup.hex(),
feature_cleanup.dll(),
feature_cleanup.fileref(),
feature_cleanup.synonyms(),
feature_cleanup.crash(),
]
if cleanup_urls:
cleanup_functions.append(feature_cleanup.url())
self.extraction_pipeline = Pipeline([
(
"bug_extractor",
bug_features.BugExtractor(feature_extractors,
cleanup_functions,
rollback=True),
),
(
"union",
ColumnTransformer([
("text", self.text_vectorizer(), "text"),
("couple_data", DictVectorizer(), "couple_data"),
]),
),
])
self.clf = XGBClassifier(n_jobs=utils.get_physical_cpu_count())
def __init__(self, lemmatization=False):
CommitModel.__init__(self, lemmatization)
self.calculate_importance = False
self.sampler = RandomUnderSampler(random_state=0)
feature_extractors = [
commit_features.files_modified_num(),
commit_features.test_added(),
commit_features.added(),
commit_features.deleted(),
commit_features.test_deleted(),
commit_features.author_experience(),
commit_features.author_experience_90_days(),
commit_features.reviewer_experience(),
commit_features.reviewer_experience_90_days(),
commit_features.components_touched_prev(),
commit_features.components_touched_prev_90_days(),
commit_features.files_touched_prev(),
commit_features.files_touched_prev_90_days(),
commit_features.types(),
commit_features.components(),
commit_features.number_of_reviewers(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"commit_extractor",
commit_features.CommitExtractor(
feature_extractors, cleanup_functions
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("desc", self.text_vectorizer(), "desc"),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
def __init__(self, lemmatization=False):
BugModel.__init__(self, lemmatization)
self.sampler = RandomUnderSampler(random_state=0)
self.calculate_importance = False
feature_extractors = [
bug_features.has_str(),
bug_features.has_regression_range(),
bug_features.severity(),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.product(),
bug_features.component(),
bug_features.num_words_title(),
bug_features.num_words_comments(),
bug_features.keywords(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"bug_extractor",
bug_features.BugExtractor(
feature_extractors, cleanup_functions, rollback=True
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(), "title"),
("comments", self.text_vectorizer(), "comments"),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
def __init__(self, cleanup_urls=True):
self.cleanup_functions = [
feature_cleanup.responses(),
feature_cleanup.hex(),
feature_cleanup.dll(),
feature_cleanup.fileref(),
feature_cleanup.synonyms(),
feature_cleanup.crash(),
]
if cleanup_urls:
self.cleanup_functions.append(feature_cleanup.url())
def __init__(self,
cleanup_urls=True,
nltk_tokenizer=False,
confidence_threshold=0.8):
self.cleanup_functions = [
feature_cleanup.responses(),
feature_cleanup.hex(),
feature_cleanup.dll(),
feature_cleanup.fileref(),
feature_cleanup.synonyms(),
feature_cleanup.crash(),
]
if cleanup_urls:
self.cleanup_functions.append(feature_cleanup.url())
self.nltk_tokenizer = nltk_tokenizer
self.confidence_threshold = confidence_threshold
def __init__(self, lemmatization=False):
BugModel.__init__(self, lemmatization)
self.sampler = RandomUnderSampler(random_state=0)
feature_extractors = [
bug_features.has_regression_range(),
bug_features.severity(),
bug_features.keywords({"stepswanted"}),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline([
(
"bug_extractor",
bug_features.BugExtractor(feature_extractors,
cleanup_functions),
),
(
"union",
ColumnTransformer([
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(), "title"),
("comments", self.text_vectorizer(), "comments"),
]),
),
])
self.clf = xgboost.XGBClassifier(n_jobs=utils.get_physical_cpu_count())
self.clf.set_params(predictor="cpu_predictor")
def __init__(self,
training_size=14000,
lemmatization=False,
cleanup_urls=True):
self.num_duplicates = training_size // 2
self.num_nondups_nondups = self.num_dup_nondups = training_size // 4
BugCoupleModel.__init__(self, lemmatization)
self.calculate_importance = False
feature_extractors = [bug_features.is_same_product()]
cleanup_functions = [
feature_cleanup.responses(),
feature_cleanup.hex(),
feature_cleanup.dll(),
feature_cleanup.fileref(),
feature_cleanup.synonyms(),
feature_cleanup.crash(),
]
if cleanup_urls:
cleanup_functions.append(feature_cleanup.url())
self.extraction_pipeline = Pipeline([
(
"bug_extractor",
bug_features.BugExtractor(feature_extractors,
cleanup_functions),
),
(
"union",
ColumnTransformer([("text", self.text_vectorizer(), "text")]),
),
])
self.clf = LinearSVCWithLabelEncoding(LinearSVC())
def __init__(self, lemmatization=False):
IssueModel.__init__(self,
owner="webcompat",
repo="web-bugs",
lemmatization=lemmatization)
self.calculate_importance = False
feature_extractors = []
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline([
(
"issue_extractor",
issue_features.IssueExtractor(feature_extractors,
cleanup_functions,
rollback=True),
),
(
"union",
ColumnTransformer([
("title", self.text_vectorizer(min_df=0.0001), "title"),
(
"first_comment",
self.text_vectorizer(min_df=0.0001),
"first_comment",
),
]),
),
])
self.clf = xgboost.XGBClassifier(n_jobs=utils.get_physical_cpu_count())
self.clf.set_params(predictor="cpu_predictor")
def __init__(
self,
cleanup_urls=True,
nltk_tokenizer=False,
confidence_threshold=0.8,
end_to_end=False,
):
self.cleanup_functions = [
feature_cleanup.responses(),
feature_cleanup.hex(),
feature_cleanup.dll(),
feature_cleanup.fileref(),
feature_cleanup.synonyms(),
feature_cleanup.crash(),
]
if cleanup_urls:
self.cleanup_functions.append(feature_cleanup.url())
self.nltk_tokenizer = nltk_tokenizer
self.confidence_threshold = confidence_threshold
self.duplicatemodel = (DuplicateModel.load("duplicatemodel")
if end_to_end else None)
def __init__(self, lemmatization=False):
BugModel.__init__(self, lemmatization)
self.sampler = InstanceHardnessThreshold(random_state=0)
feature_extractors = [
bug_features.has_str(),
bug_features.has_regression_range(),
bug_features.severity(),
bug_features.keywords(),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.title(),
bug_features.product(),
bug_features.component(),
bug_features.is_mozillian(),
bug_features.bug_reporter(),
bug_features.blocked_bugs_number(),
bug_features.priority(),
bug_features.has_cve_in_alias(),
bug_features.comment_count(),
bug_features.comment_length(),
bug_features.reporter_experience(),
bug_features.number_of_bug_dependencies(),
]
cleanup_functions = [
feature_cleanup.url(),
feature_cleanup.fileref(),
feature_cleanup.hex(),
feature_cleanup.dll(),
feature_cleanup.synonyms(),
feature_cleanup.crash(),
]
self.extraction_pipeline = Pipeline(
[
(
"bug_extractor",
bug_features.BugExtractor(
feature_extractors,
cleanup_functions,
rollback=True,
rollback_when=self.rollback,
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.0001), "title"),
(
"comments",
self.text_vectorizer(min_df=0.0001),
"comments",
),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
def __init__(self,
lemmatization=False,
historical=False,
rca_subcategories_enabled=False):
BugModel.__init__(self, lemmatization)
self.calculate_importance = False
self.rca_subcategories_enabled = rca_subcategories_enabled
# should we consider only the main category or all sub categories
self.RCA_TYPES = (RCA_SUBCATEGORIES + RCA_CATEGORIES
if rca_subcategories_enabled else RCA_CATEGORIES)
self.RCA_LIST = sorted(set(self.RCA_TYPES))
feature_extractors = [
bug_features.has_str(),
bug_features.severity(),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
# Ignore whiteboards that would make the ML completely skewed
# bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.blocked_bugs_number(),
bug_features.ever_affected(),
bug_features.affected_then_unaffected(),
bug_features.product(),
bug_features.component(),
]
cleanup_functions = [
feature_cleanup.url(),
feature_cleanup.fileref(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline([
(
"bug_extractor",
bug_features.BugExtractor(feature_extractors,
cleanup_functions),
),
(
"union",
ColumnTransformer([
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.001), "title"),
(
"first_comment",
self.text_vectorizer(min_df=0.001),
"first_comment",
),
(
"comments",
self.text_vectorizer(min_df=0.001),
"comments",
),
]),
),
])
self.clf = OneVsRestClassifier(xgboost.XGBClassifier(n_jobs=16))
def __init__(self, lemmatization=False):
BugModel.__init__(self, lemmatization)
self.calculate_importance = False
self.sampler = InstanceHardnessThreshold(random_state=0)
feature_extractors = [
bug_features.has_str(),
bug_features.has_regression_range(),
bug_features.severity(),
bug_features.keywords(),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.title(),
bug_features.product(),
bug_features.component(),
bug_features.is_mozillian(),
bug_features.bug_reporter(),
bug_features.blocked_bugs_number(),
bug_features.priority(),
bug_features.has_cve_in_alias(),
bug_features.comment_count(),
bug_features.comment_length(),
bug_features.reporter_experience(),
bug_features.number_of_bug_dependencies(),
]
cleanup_functions = [
feature_cleanup.url(),
feature_cleanup.fileref(),
feature_cleanup.hex(),
feature_cleanup.dll(),
feature_cleanup.synonyms(),
feature_cleanup.crash(),
]
self.extraction_pipeline = Pipeline([
(
"bug_extractor",
bug_features.BugExtractor(
feature_extractors,
cleanup_functions,
rollback=True,
rollback_when=self.rollback,
),
),
(
"union",
ColumnTransformer([
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.0001), "title"),
(
"comments",
self.text_vectorizer(min_df=0.0001),
"comments",
),
]),
),
])
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
def __init__(self, lemmatization=False, bug_data=False):
CommitModel.__init__(self, lemmatization, bug_data)
self.calculate_importance = False
self.sampler = RandomUnderSampler(random_state=0)
feature_extractors = [
commit_features.source_code_files_modified_num(),
commit_features.other_files_modified_num(),
commit_features.test_files_modified_num(),
commit_features.source_code_file_size(),
commit_features.other_file_size(),
commit_features.test_file_size(),
commit_features.source_code_added(),
commit_features.other_added(),
commit_features.test_added(),
commit_features.source_code_deleted(),
commit_features.other_deleted(),
commit_features.test_deleted(),
commit_features.author_experience(),
commit_features.reviewer_experience(),
commit_features.reviewers_num(),
commit_features.component_touched_prev(),
commit_features.directory_touched_prev(),
commit_features.file_touched_prev(),
commit_features.types(),
commit_features.components(),
commit_features.directories(),
commit_features.files(),
]
if bug_data:
feature_extractors += [
bug_features.product(),
bug_features.component(),
bug_features.severity(),
bug_features.priority(),
bug_features.has_crash_signature(),
bug_features.has_regression_range(),
bug_features.whiteboard(),
bug_features.keywords(),
bug_features.number_of_bug_dependencies(),
bug_features.blocked_bugs_number(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"commit_extractor",
commit_features.CommitExtractor(
feature_extractors, cleanup_functions
),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("desc", self.text_vectorizer(), "desc"),
]
),
),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=utils.get_physical_cpu_count())
self.clf.set_params(predictor="cpu_predictor")
def __init__(self, lemmatization=False, historical=False):
BugModel.__init__(self, lemmatization)
self.sampler = BorderlineSMOTE(random_state=0)
feature_extractors = [
bug_features.has_str(),
bug_features.severity(),
# Ignore keywords that would make the ML completely skewed
# (we are going to use them as 100% rules in the evaluation phase).
bug_features.keywords(set(keyword_dict.keys())),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.title(),
bug_features.blocked_bugs_number(),
bug_features.ever_affected(),
bug_features.affected_then_unaffected(),
bug_features.product(),
bug_features.component(),
]
cleanup_functions = [
feature_cleanup.url(),
feature_cleanup.fileref(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline(
[
(
"bug_extractor",
bug_features.BugExtractor(feature_extractors, cleanup_functions),
),
(
"union",
ColumnTransformer(
[
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.001), "title"),
(
"first_comment",
self.text_vectorizer(min_df=0.001),
"first_comment",
),
(
"comments",
self.text_vectorizer(min_df=0.001),
"comments",
),
]
),
),
]
)
self.clf = OneVsRestClassifier(xgboost.XGBClassifier(n_jobs=16))
def __init__(self, lemmatization=False, historical=False):
BugModel.__init__(self, lemmatization)
self.sampler = BorderlineSMOTE(random_state=0)
feature_extractors = [
bug_features.has_str(),
bug_features.severity(),
# Ignore keywords that would make the ML completely skewed
# (we are going to use them as 100% rules in the evaluation phase).
bug_features.keywords(
{"regression", "talos-regression", "feature"}),
bug_features.is_coverity_issue(),
bug_features.has_crash_signature(),
bug_features.has_url(),
bug_features.has_w3c_url(),
bug_features.has_github_url(),
bug_features.whiteboard(),
bug_features.patches(),
bug_features.landings(),
bug_features.title(),
bug_features.blocked_bugs_number(),
bug_features.ever_affected(),
bug_features.affected_then_unaffected(),
bug_features.product(),
bug_features.component(),
]
if historical:
feature_extractors.append(bug_features.had_severity_enhancement())
cleanup_functions = [
feature_cleanup.url(),
feature_cleanup.fileref(),
feature_cleanup.synonyms(),
]
self.extraction_pipeline = Pipeline([
(
"bug_extractor",
bug_features.BugExtractor(feature_extractors,
cleanup_functions),
),
(
"union",
ColumnTransformer([
("data", DictVectorizer(), "data"),
("title", self.text_vectorizer(min_df=0.001), "title"),
(
"first_comment",
self.text_vectorizer(min_df=0.001),
"first_comment",
),
(
"comments",
self.text_vectorizer(min_df=0.001),
"comments",
),
]),
),
])
self.clf = xgboost.XGBClassifier(n_jobs=16)
self.clf.set_params(predictor="cpu_predictor")
from nltk.corpus import stopwords
from nltk.stem.porter import PorterStemmer
except ImportError:
raise ImportError(OPT_MSG_MISSING)
nltk.download("stopwords")
REPORTERS_TO_IGNORE = {
"*****@*****.**", "*****@*****.**"
}
cleanup_functions = [
feature_cleanup.responses(),
feature_cleanup.hex(),
feature_cleanup.dll(),
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
feature_cleanup.crash(),
]
# A map from bug ID to its duplicate IDs
duplicates = defaultdict(set)
all_ids = set(bug["id"] for bug in bugzilla.get_bugs()
if bug["creator"] not in REPORTERS_TO_IGNORE
and "dupeme" not in bug["keywords"])
for bug in bugzilla.get_bugs():
dupes = [entry for entry in bug["duplicates"] if entry in all_ids]
if bug["dupe_of"] in all_ids:
dupes.append(bug["dupe_of"])
def test_BugExtractor():
BugExtractor([has_str(), has_url()], [fileref(), url()])
with pytest.raises(AssertionError):
BugExtractor([has_str(), has_str()], [fileref(), url()])
with pytest.raises(AssertionError):
BugExtractor([has_str(), has_url()], [fileref(), fileref()])
def __init__(
self,
lemmatization: bool = False,
interpretable: bool = True,
use_finder: bool = False,
exclude_finder: bool = True,
finder_regressions_only: bool = False,
) -> None:
CommitModel.__init__(self, lemmatization)
self.training_dbs += [BUG_INTRODUCING_COMMITS_DB, bugzilla.BUGS_DB]
if finder_regressions_only:
self.training_dbs.append(BUG_FIXING_COMMITS_DB)
self.store_dataset = True
self.sampler = RandomUnderSampler(random_state=0)
self.use_finder = use_finder
self.exclude_finder = exclude_finder
assert (
use_finder ^ exclude_finder
), "Using both use_finder and exclude_finder option does not make a lot of sense"
self.finder_regressions_only = finder_regressions_only
feature_extractors = [
commit_features.source_code_file_size(),
commit_features.other_file_size(),
commit_features.test_file_size(),
commit_features.source_code_added(),
commit_features.other_added(),
commit_features.test_added(),
commit_features.source_code_deleted(),
commit_features.other_deleted(),
commit_features.test_deleted(),
commit_features.author_experience(),
commit_features.reviewer_experience(),
commit_features.reviewers_num(),
commit_features.component_touched_prev(),
commit_features.directory_touched_prev(),
commit_features.file_touched_prev(),
commit_features.types(),
commit_features.files(),
commit_features.components(),
commit_features.components_modified_num(),
commit_features.directories(),
commit_features.directories_modified_num(),
commit_features.source_code_files_modified_num(),
commit_features.other_files_modified_num(),
commit_features.test_files_modified_num(),
commit_features.functions_touched_num(),
commit_features.functions_touched_size(),
commit_features.source_code_file_metrics(),
]
cleanup_functions = [
feature_cleanup.fileref(),
feature_cleanup.url(),
feature_cleanup.synonyms(),
]
column_transformers = [("data", DictVectorizer(), "data")]
if not interpretable:
column_transformers.append(
("desc", self.text_vectorizer(min_df=0.0001), "desc")
)
self.extraction_pipeline = Pipeline(
[
(
"commit_extractor",
commit_features.CommitExtractor(
feature_extractors, cleanup_functions
),
),
("union", ColumnTransformer(column_transformers)),
]
)
self.clf = xgboost.XGBClassifier(n_jobs=utils.get_physical_cpu_count())
self.clf.set_params(predictor="cpu_predictor")