def test_query(query_strategy,
n_features=50,
n_sample=100,
n_instances_list=[0, 1, 5, 50],
n_train_idx=[0, 1, 5, 50]):
classifier = get_model("rf")
if query_strategy == "cluster":
data_fp = os.path.join("test", "demo_data", "generic.csv")
texts = ASReviewData.from_file(data_fp).texts
while len(texts) < n_features:
texts = np.append(texts, texts)
print(len(texts))
# texts.extend(texts)
texts = texts[:n_features]
query_model = get_query_model(query_strategy,
texts=texts,
update_interval=None,
cluster_size=int(n_sample / 3))
assert isinstance(query_model.param, dict)
else:
query_model = get_query_model(query_strategy)
X = np.random.rand(n_sample, n_features)
y = np.concatenate((np.zeros(n_sample // 2), np.ones(n_sample // 2)),
axis=0)
print(X.shape, y.shape)
order = np.random.permutation(n_sample)
print(order.shape)
X = X[order]
y = y[order]
sources = query_strategy.split('_')
classifier.fit(X, y)
assert isinstance(query_model.param, dict)
assert query_model.name == query_strategy
for n_instances in n_instances_list:
for n_train in n_train_idx:
shared = {"query_src": {}, "current_queries": {}}
train_idx = np.random.choice(np.arange(n_sample),
n_train,
replace=False)
pool_idx = np.delete(np.arange(n_sample), train_idx)
query_idx, X_query = query_model.query(X, classifier, pool_idx,
n_instances, shared)
check_integrity(query_idx, X_query, X, pool_idx, shared,
n_instances, sources)
def test_features(feature_extraction, split_ta):
embedding_fp = os.path.join("test", "demo_data", "generic.vec")
data_fp = os.path.join("test", "demo_data", "generic.csv")
as_data = ASReviewData.from_file(data_fp)
texts = as_data.texts
if feature_extraction.startswith("embedding-"):
model = get_feature_model(feature_extraction,
split_ta=split_ta,
embedding_fp=embedding_fp)
else:
model = get_feature_model(feature_extraction, split_ta=split_ta)
X = model.fit_transform(texts,
titles=as_data.title,
abstracts=as_data.abstract)
assert X.shape[0] == len(as_data.title)
assert X.shape[1] > 0
assert isinstance(model.param, dict)
assert model.name == feature_extraction
def test_csv_write_data():
fp_in = Path("test", "demo_data", "csv_example_with_labels.csv")
fp_out = Path("test", "out_data", "csv_out_example.csv")
asr_data = ASReviewData.from_file(fp_in)
asr_data.to_csv(fp_out, labels=[0, 1, 0, 1, 0, 1])
def get_reviewer(dataset,
mode='oracle',
model=DEFAULT_MODEL,
query_strategy=DEFAULT_QUERY_STRATEGY,
balance_strategy=DEFAULT_BALANCE_STRATEGY,
n_instances=DEFAULT_N_INSTANCES,
n_papers=None,
n_queries=None,
embedding_fp=None,
verbose=0,
prior_included=None,
prior_excluded=None,
n_prior_included=DEFAULT_N_PRIOR_INCLUDED,
n_prior_excluded=DEFAULT_N_PRIOR_EXCLUDED,
config_file=None,
log_file=None,
model_param=None,
query_param=None,
balance_param=None,
abstract_only=False,
**kwargs):
""" Get a review object from arguments. See __main__.py for a description
Of the arguments.
"""
# Find the URL of the datasets if the dataset is an example dataset.
if dataset in DEMO_DATASETS.keys():
dataset = DEMO_DATASETS[dataset]
cli_settings = ASReviewSettings(model=model,
n_instances=n_instances,
n_queries=n_queries,
n_papers=n_papers,
n_prior_included=n_prior_included,
n_prior_excluded=n_prior_excluded,
query_strategy=query_strategy,
balance_strategy=balance_strategy,
mode=mode,
data_fp=dataset,
abstract_only=abstract_only)
cli_settings.from_file(config_file)
if log_file is not None:
with open_logger(log_file) as logger:
if logger.is_empty():
logger.add_settings(cli_settings)
settings = logger.settings
else:
settings = cli_settings
logger = None
if n_queries is not None:
settings.n_queries = n_queries
if n_papers is not None:
settings.n_papers = n_papers
if model_param is not None:
settings.model_param = model_param
if query_param is not None:
settings.query_param = query_param
if balance_param is not None:
settings.balance_param = balance_param
model = settings.model
# Check if mode is valid
if mode in AVAILABLE_REVIEW_CLASSES:
logging.info(f"Start review in '{mode}' mode.")
else:
raise ValueError(f"Unknown mode '{mode}'.")
logging.debug(settings)
as_data = ASReviewData.from_file(dataset,
abstract_only=settings.abstract_only)
_, texts, labels = as_data.get_data()
if as_data.final_labels is not None:
with open_logger(log_file) as logger:
logger.set_final_labels(as_data.final_labels)
model_class = get_model_class(model)
model_inst = model_class(param=settings.model_param,
embedding_fp=embedding_fp)
X, y = model_inst.get_Xy(texts, labels)
model_fn = model_inst.model()
settings.fit_kwargs = model_inst.fit_kwargs()
settings.query_kwargs = {}
# Pick query strategy
query_fn, query_str = get_query_with_settings(settings)
logging.info(f"Query strategy: {query_str}")
train_data_fn, train_method = get_balance_with_settings(settings)
logging.info(f"Using {train_method} method to obtain training data.")
# Initialize the review class.
if mode == "simulate":
reviewer = ReviewSimulate(X,
y,
model=model_fn,
query_strategy=query_fn,
train_data_fn=train_data_fn,
n_papers=settings.n_papers,
n_instances=settings.n_instances,
n_queries=settings.n_queries,
verbose=verbose,
prior_included=prior_included,
prior_excluded=prior_excluded,
n_prior_included=settings.n_prior_included,
n_prior_excluded=settings.n_prior_excluded,
fit_kwargs=settings.fit_kwargs,
balance_kwargs=settings.balance_kwargs,
query_kwargs=settings.query_kwargs,
log_file=log_file,
final_labels=as_data.final_labels,
**kwargs)
elif mode == "oracle":
reviewer = ReviewOracle(X,
model=model_fn,
query_strategy=query_fn,
as_data=as_data,
train_data_fn=train_data_fn,
n_papers=settings.n_papers,
n_instances=settings.n_instances,
n_queries=settings.n_queries,
verbose=verbose,
prior_included=prior_included,
prior_excluded=prior_excluded,
fit_kwargs=settings.fit_kwargs,
balance_kwargs=settings.balance_kwargs,
query_kwargs=settings.query_kwargs,
log_file=log_file,
**kwargs)
elif mode == "minimal":
reviewer = MinimalReview(X,
model=model_fn,
query_strategy=query_fn,
train_data_fn=train_data_fn,
n_papers=settings.n_papers,
n_instances=settings.n_instances,
n_queries=settings.n_queries,
verbose=verbose,
prior_included=prior_included,
prior_excluded=prior_excluded,
fit_kwargs=settings.fit_kwargs,
balance_kwargs=settings.balance_kwargs,
query_kwargs=settings.query_kwargs,
log_file=log_file,
**kwargs)
else:
raise ValueError("Error finding mode, should never come here...")
return reviewer
def get_reviewer(dataset,
mode='oracle',
model=DEFAULT_MODEL,
query_strategy=DEFAULT_QUERY_STRATEGY,
balance_strategy=DEFAULT_BALANCE_STRATEGY,
feature_extraction=DEFAULT_FEATURE_EXTRACTION,
n_instances=DEFAULT_N_INSTANCES,
n_papers=None,
n_queries=None,
embedding_fp=None,
verbose=0,
prior_included=None,
prior_excluded=None,
n_prior_included=DEFAULT_N_PRIOR_INCLUDED,
n_prior_excluded=DEFAULT_N_PRIOR_EXCLUDED,
config_file=None,
log_file=None,
model_param=None,
query_param=None,
balance_param=None,
feature_param=None,
abstract_only=False,
extra_dataset=[],
**kwargs
):
""" Get a review object from arguments. See __main__.py for a description
Of the arguments.
"""
# Find the URL of the datasets if the dataset is an example dataset.
if dataset in DEMO_DATASETS.keys():
dataset = DEMO_DATASETS[dataset]
cli_settings = ASReviewSettings(
model=model, n_instances=n_instances, n_queries=n_queries,
n_papers=n_papers, n_prior_included=n_prior_included,
n_prior_excluded=n_prior_excluded, query_strategy=query_strategy,
balance_strategy=balance_strategy,
feature_extraction=feature_extraction,
mode=mode, data_fp=dataset,
abstract_only=abstract_only)
cli_settings.from_file(config_file)
if log_file is not None:
with open_logger(log_file) as logger:
if logger.is_empty():
logger.add_settings(cli_settings)
settings = logger.settings
else:
settings = cli_settings
logger = None
if n_queries is not None:
settings.n_queries = n_queries
if n_papers is not None:
settings.n_papers = n_papers
if model_param is not None:
settings.model_param = model_param
if query_param is not None:
settings.query_param = query_param
if balance_param is not None:
settings.balance_param = balance_param
if feature_param is not None:
settings.feature_param = feature_param
# Check if mode is valid
if mode in AVAILABLE_REVIEW_CLASSES:
logging.info(f"Start review in '{mode}' mode.")
else:
raise ValueError(f"Unknown mode '{mode}'.")
logging.debug(settings)
as_data = ASReviewData.from_file(dataset, extra_dataset=extra_dataset,
abstract_only=settings.abstract_only)
texts = as_data.texts
y = as_data.labels
data_prior_included, data_prior_excluded = as_data.get_priors()
if len(data_prior_included) != 0:
if prior_included is None:
prior_included = []
prior_included.extend(data_prior_included.tolist())
if len(data_prior_excluded) != 0:
if prior_excluded is None:
prior_excluded = []
prior_excluded.extend(data_prior_excluded.tolist())
if as_data.final_labels is not None:
with open_logger(log_file) as logger:
logger.set_final_labels(as_data.final_labels)
train_model = get_model(settings.model, **settings.model_param)
query_model = get_query_model(settings.query_strategy,
**settings.query_param)
balance_model = get_balance_model(settings.balance_strategy,
**settings.balance_param)
feature_model = get_feature_model(settings.feature_extraction,
**settings.feature_param)
X = feature_model.fit_transform(texts, as_data.title, as_data.abstract)
if train_model.name.startswith("lstm-"):
train_model.embedding_matrix = feature_model.get_embedding_matrix(
texts, embedding_fp)
# Initialize the review class.
if mode == "simulate":
reviewer = ReviewSimulate(
X, y,
model=train_model,
query_model=query_model,
balance_model=balance_model,
feature_model=feature_model,
n_papers=settings.n_papers,
n_instances=settings.n_instances,
n_queries=settings.n_queries,
verbose=verbose,
prior_included=prior_included,
prior_excluded=prior_excluded,
n_prior_included=settings.n_prior_included,
n_prior_excluded=settings.n_prior_excluded,
log_file=log_file,
final_labels=as_data.final_labels,
data_fp=dataset,
**kwargs)
elif mode == "oracle":
reviewer = ReviewOracle(
X,
model=train_model,
query_model=query_model,
balance_model=balance_model,
feature_model=feature_model,
as_data=as_data,
n_papers=settings.n_papers,
n_instances=settings.n_instances,
n_queries=settings.n_queries,
verbose=verbose,
prior_included=prior_included,
prior_excluded=prior_excluded,
log_file=log_file,
data_fp=dataset,
**kwargs)
elif mode == "minimal":
reviewer = MinimalReview(
X,
model=model,
query_model=query_model,
balance_model=balance_model,
feature_model=feature_model,
n_papers=settings.n_papers,
n_instances=settings.n_instances,
n_queries=settings.n_queries,
verbose=verbose,
prior_included=prior_included,
prior_excluded=prior_excluded,
log_file=log_file,
data_fp=dataset,
**kwargs)
else:
raise ValueError("Error finding mode, should never come here...")
return reviewer
for text in texts:
n_numbers.append(sum(c.isdigit() for c in text))
return np.array(n_numbers, dtype=int)
if __name__ == "__main__":
data_dir = sys.argv[1]
data_fp = sys.argv[2]
analysis = Analysis.from_dir(data_dir)
ttd = analysis.avg_time_to_discovery()
ttd_order = sorted(ttd, key=lambda x: ttd[x])
# for idx in ttd_order:
# print(f"{idx}: {ttd[idx]}")
as_data = ASReviewData.from_file(data_fp)
n_abstract_missing = 0
n_missing_included = 0
for i, abstract in enumerate(as_data.abstract):
if len(abstract) < 10:
n_abstract_missing += 1
if as_data.labels[i] == 1:
n_missing_included += 1
n_paper = len(as_data.abstract)
n_included = np.sum(as_data.labels)
print(f"Number of abstracts missing: {n_abstract_missing}/{n_paper}")
print(
f"Number of included abstracts missing: {n_missing_included}/{n_included}"
)
for idx in ttd_order[-10:]:
def get_reviewer(dataset,
mode='oracle',
model=DEFAULT_MODEL,
query_strategy=DEFAULT_QUERY_STRATEGY,
balance_strategy=DEFAULT_BALANCE_STRATEGY,
n_instances=DEFAULT_N_INSTANCES,
n_queries=1,
embedding_fp=None,
verbose=1,
prior_included=None,
prior_excluded=None,
n_prior_included=DEFAULT_N_PRIOR_INCLUDED,
n_prior_excluded=DEFAULT_N_PRIOR_EXCLUDED,
config_file=None,
src_log_fp=None,
**kwargs):
# Find the URL of the datasets if the dataset is an example dataset.
if dataset in DEMO_DATASETS.keys():
dataset = DEMO_DATASETS[dataset]
if src_log_fp is not None:
logger = Logger(log_fp=src_log_fp)
settings = logger.settings
else:
logger = None
settings = ASReviewSettings(model=model,
n_instances=n_instances,
n_queries=n_queries,
n_prior_included=n_prior_included,
n_prior_excluded=n_prior_excluded,
query_strategy=query_strategy,
balance_strategy=balance_strategy,
mode=mode,
data_fp=dataset)
settings.from_file(config_file)
model = settings.model
if model in ["lstm_base", "lstm_pool"]:
base_model = "RNN"
else:
base_model = "other"
# Check if mode is valid
if mode in AVAILABLE_REVIEW_CLASSES:
if verbose:
print(f"Start review in '{mode}' mode.")
else:
raise ValueError(f"Unknown mode '{mode}'.")
print(f"Model: '{model}'")
# if the provided file is a pickle file
if is_pickle(dataset):
with open(dataset, 'rb') as f:
data_obj = pickle.load(f)
if isinstance(data_obj, tuple) and len(data_obj) == 3:
X, y, embedding_matrix = data_obj
elif isinstance(data_obj, tuple) and len(data_obj) == 4:
X, y, embedding_matrix, _ = data_obj
else:
raise ValueError("Incorrect pickle object.")
else:
as_data = ASReviewData.from_file(dataset)
_, texts, labels = as_data.get_data()
# get the model
if base_model == "RNN":
if embedding_fp is None:
embedding_fp = Path(get_data_home(),
EMBEDDING_EN["name"]).expanduser()
if not embedding_fp.exists():
print("Warning: will start to download large "
"embedding file in 10 seconds.")
time.sleep(10)
download_embedding(verbose=verbose)
# create features and labels
X, word_index = text_to_features(texts)
y = labels
embedding = load_embedding(embedding_fp, word_index=word_index)
embedding_matrix = sample_embedding(embedding, word_index)
elif model.lower() in ['nb', 'svc', 'svm']:
from sklearn.pipeline import Pipeline
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import CountVectorizer
text_clf = Pipeline([('vect', CountVectorizer()),
('tfidf', TfidfTransformer())])
X = text_clf.fit_transform(texts)
y = labels
settings.fit_kwargs = {}
settings.query_kwargs = {}
if base_model == 'RNN':
if model == "lstm_base":
model_kwargs = lstm_base_model_defaults(settings, verbose)
create_lstm_model = create_lstm_base_model
elif model == "lstm_pool":
model_kwargs = lstm_pool_model_defaults(settings, verbose)
create_lstm_model = create_lstm_pool_model
else:
raise ValueError(f"Unknown model {model}")
settings.fit_kwargs = lstm_fit_defaults(settings, verbose)
settings.query_kwargs['verbose'] = verbose
# create the model
model = KerasClassifier(create_lstm_model(
embedding_matrix=embedding_matrix, **model_kwargs),
verbose=verbose)
elif model.lower() in ['nb']:
from asreview.models import create_nb_model
model = create_nb_model()
elif model.lower() in ['svm', 'svc']:
from asreview.models import create_svc_model
model = create_svc_model()
else:
raise ValueError('Model not found.')
# Pick query strategy
query_fn, query_str = get_query_strategy(settings)
if verbose:
print(f"Query strategy: {query_str}")
train_data_fn, train_method = get_balance_strategy(settings)
if verbose:
print(f"Using {train_method} method to obtain training data.")
# Initialize the review class.
if mode == "simulate":
reviewer = ReviewSimulate(X,
y,
model=model,
query_strategy=query_fn,
train_data_fn=train_data_fn,
n_instances=settings.n_instances,
n_queries=settings.n_queries,
verbose=verbose,
prior_included=prior_included,
prior_excluded=prior_excluded,
n_prior_included=settings.n_prior_included,
n_prior_excluded=settings.n_prior_excluded,
fit_kwargs=settings.fit_kwargs,
balance_kwargs=settings.balance_kwargs,
query_kwargs=settings.query_kwargs,
logger=logger,
**kwargs)
elif mode == "oracle":
reviewer = ReviewOracle(X,
model=model,
query_strategy=query_fn,
as_data=as_data,
train_data_fn=train_data_fn,
n_instances=settings.n_instances,
n_queries=settings.n_queries,
verbose=verbose,
prior_included=prior_included,
prior_excluded=prior_excluded,
fit_kwargs=settings.fit_kwargs,
balance_kwargs=settings.balance_kwargs,
query_kwargs=settings.query_kwargs,
logger=logger,
**kwargs)
elif mode == "minimal":
reviewer = MinimalReview(X,
model=model,
query_strategy=query_fn,
train_data_fn=train_data_fn,
n_instances=settings.n_instances,
n_queries=settings.n_queries,
verbose=verbose,
prior_included=prior_included,
prior_excluded=prior_excluded,
fit_kwargs=settings.fit_kwargs,
balance_kwargs=settings.balance_kwargs,
query_kwargs=settings.query_kwargs,
logger=logger,
**kwargs)
else:
raise ValueError("Error finding mode, should never come here...")
reviewer._logger.add_settings(settings)
return reviewer
def test_csv_write_data():
fp_in = Path("test", "demo_data", "generic_labels.csv")
fp_out = Path("test", "out_data", "generic_out.csv")
asr_data = ASReviewData.from_file(fp_in)
asr_data.to_csv(fp_out, labels=[0, 1, 0, 1, 0, 1])