def add_dataset_to_project(project_id, file_name):
"""Add file path to the project file.
Add file to data subfolder and fill the pool of iteration 0.
"""
project_file_path = get_project_file_path(project_id)
fp_lock = get_lock_path(project_id)
with SQLiteLock(fp_lock, blocking=True, lock_name="active"):
# open the projects file
with open(project_file_path, "r") as f_read:
project_dict = json.load(f_read)
# add path to dict (overwrite if already exists)
project_dict["dataset_path"] = file_name
with open(project_file_path, "w") as f_write:
json.dump(project_dict, f_write)
# fill the pool of the first iteration
pool_indices = read_data(project_id).record_ids
np.random.shuffle(pool_indices)
write_pool(project_id, pool_indices.tolist())
# make a empty qeue for the items to label
write_label_history(project_id, [])
def move_label_from_labeled_to_pool(project_id, paper_i):
# load the papers from the pool
pool_list = read_pool(project_id)
# Add the paper to the reviewed papers.
labeled_list = read_label_history(project_id)
labeled_list_new = []
for item_id, item_label in labeled_list:
item_id = int(item_id)
item_label = int(item_label)
paper_i = int(paper_i)
if paper_i == item_id:
pool_list.append(item_id)
else:
labeled_list_new.append([item_id, item_label])
# write the papers to the label dataset
write_pool(project_id, pool_list)
# load the papers from the pool
write_label_history(project_id, labeled_list_new)
def add_dataset_to_project(project_id, file_name):
"""Add file path to the project file.
Add file to data subfolder and fill the pool of iteration 0.
"""
project_file_path = get_project_file_path(project_id)
# clean temp project files
clean_project_tmp_files(project_id)
with SQLiteLock(
get_lock_path(project_id),
blocking=True,
lock_name="active",
project_id=project_id
):
# open the projects file
with open(project_file_path, "r") as f_read:
project_dict = json.load(f_read)
# add path to dict (overwrite if already exists)
project_dict["dataset_path"] = file_name
with open(project_file_path, "w") as f_write:
json.dump(project_dict, f_write)
# fill the pool of the first iteration
as_data = read_data(project_id)
if as_data.labels is not None:
unlabeled = np.where(as_data.labels == LABEL_NA)[0]
pool_indices = as_data.record_ids[unlabeled]
labeled_indices = np.where(as_data.labels != LABEL_NA)[0]
label_indices = list(zip(
as_data.record_ids[labeled_indices].tolist(),
as_data.labels[labeled_indices].tolist()
))
else:
pool_indices = as_data.record_ids
label_indices = []
np.random.shuffle(pool_indices)
write_pool(project_id, pool_indices.tolist())
# make a empty qeue for the items to label
write_label_history(project_id, label_indices)
def move_label_from_pool_to_labeled(project_id, paper_i, label):
# load the papers from the pool
pool_idx = read_pool(project_id)
# Remove the paper from the pool.
try:
pool_idx.remove(int(paper_i))
except (IndexError, ValueError):
return
write_pool(project_id, pool_idx)
# Add the paper to the reviewed papers.
labeled = read_label_history(project_id)
labeled.append([int(paper_i), int(label)])
write_label_history(project_id, labeled)
def add_dataset_to_project(project_id, file_name):
"""Add file path to the project file.
Add file to data subfolder and fill the pool of iteration 0.
"""
project_file_path = get_project_file_path(project_id)
fp_lock = get_lock_path(project_id)
with SQLiteLock(fp_lock, blocking=True, lock_name="active"):
# open the projects file
with open(project_file_path, "r") as f_read:
project_dict = json.load(f_read)
# add path to dict (overwrite if already exists)
project_dict["dataset_path"] = file_name
with open(project_file_path, "w") as f_write:
json.dump(project_dict, f_write)
# fill the pool of the first iteration
as_data = read_data(project_id)
if as_data.labels is not None:
unlabeled = np.where(as_data.labels == LABEL_NA)[0]
pool_indices = as_data.record_ids[unlabeled]
label_indices_included = \
[[int(x), 1] for x in np.where(as_data.labels == 1)[0]]
label_indices_excluded = \
[[int(x), 0] for x in np.where(as_data.labels == 0)[0]]
label_indices = label_indices_included + label_indices_excluded
else:
pool_indices = as_data.record_ids
label_indices = []
np.random.shuffle(pool_indices)
write_pool(project_id, pool_indices.tolist())
# make a empty qeue for the items to label
write_label_history(project_id, label_indices)
def train_model(project_id, label_method=None):
"""Add the new labels to the review and do the modeling.
It uses a lock to ensure only one model is running at the same time.
Old results directories are deleted after 4 iterations.
It has one argument on the CLI, which is the base project directory.
"""
logging.info(f"Project {project_id} - Train a new model for project")
# get file locations
asr_kwargs_file = get_kwargs_path(project_id)
lock_file = get_lock_path(project_id)
# Lock so that only one training run is running at the same time.
# It doesn't lock the flask server/client.
with SQLiteLock(lock_file,
blocking=False,
lock_name="training",
project_id=project_id) as lock:
# If the lock is not acquired, another training instance is running.
if not lock.locked():
logging.info("Project {project_id} - "
"Cannot acquire lock, other instance running.")
return
# Lock the current state. We want to have a consistent active state.
# This does communicate with the flask backend; it prevents writing and
# reading to the same files at the same time.
with SQLiteLock(lock_file,
blocking=True,
lock_name="active",
project_id=project_id) as lock:
# Get the all labels since last run. If no new labels, quit.
new_label_history = read_label_history(project_id)
data_fp = str(get_data_file_path(project_id))
as_data = read_data(project_id)
state_file = get_state_path(project_id)
# collect command line arguments and pass them to the reviewer
with open(asr_kwargs_file, "r") as fp:
asr_kwargs = json.load(fp)
asr_kwargs['state_file'] = str(state_file)
reviewer = get_reviewer(dataset=data_fp, mode="minimal", **asr_kwargs)
with open_state(state_file) as state:
old_label_history = get_label_train_history(state)
diff_history = get_diff_history(new_label_history, old_label_history)
if len(diff_history) == 0:
logging.info(
"Project {project_id} - No new labels since last run.")
return
query_idx = np.array([x[0] for x in diff_history], dtype=int)
inclusions = np.array([x[1] for x in diff_history], dtype=int)
# Classify the new labels, train and store the results.
with open_state(state_file) as state:
reviewer.classify(query_idx,
inclusions,
state,
method=label_method)
reviewer.train()
reviewer.log_probabilities(state)
new_query_idx = reviewer.query(reviewer.n_pool()).tolist()
reviewer.log_current_query(state)
proba = state.pred_proba.tolist()
with SQLiteLock(lock_file,
blocking=True,
lock_name="active",
project_id=project_id) as lock:
current_pool = read_pool(project_id)
in_current_pool = np.zeros(len(as_data))
in_current_pool[current_pool] = 1
new_pool = [x for x in new_query_idx if in_current_pool[x]]
write_pool(project_id, new_pool)
write_proba(project_id, proba)
def train_model(project_id, label_method=None):
"""Add the new labels to the review and do the modeling.
It uses a lock to ensure only one model is running at the same time.
Old results directories are deleted after 4 iterations.
It has one argument on the CLI, which is the base project directory.
"""
logging.info(f"Project {project_id} - Train a new model for project")
# get file locations
asr_kwargs_file = get_kwargs_path(project_id)
lock_file = get_lock_path(project_id)
# Lock so that only one training run is running at the same time.
# It doesn't lock the flask server/client.
with SQLiteLock(
lock_file, blocking=False, lock_name="training",
project_id=project_id) as lock:
# If the lock is not acquired, another training instance is running.
if not lock.locked():
logging.info("Project {project_id} - "
"Cannot acquire lock, other instance running.")
return
# Lock the current state. We want to have a consistent active state.
# This does communicate with the flask backend; it prevents writing and
# reading to the same files at the same time.
with SQLiteLock(
lock_file,
blocking=True,
lock_name="active",
project_id=project_id) as lock:
# Get the all labels since last run. If no new labels, quit.
new_label_history = read_label_history(project_id)
data_fp = str(get_data_file_path(project_id))
as_data = read_data(project_id)
state_file = get_state_path(project_id)
# collect command line arguments and pass them to the reviewer
with open(asr_kwargs_file, "r") as fp:
asr_kwargs = json.load(fp)
try:
del asr_kwargs["abstract_only"]
except KeyError:
pass
asr_kwargs['state_file'] = str(state_file)
reviewer = get_reviewer(dataset=data_fp, mode="minimal", **asr_kwargs)
with open_state(state_file) as state:
old_label_history = _get_label_train_history(state)
diff_history = _get_diff_history(new_label_history, old_label_history)
if len(diff_history) == 0:
logging.info(
"Project {project_id} - No new labels since last run.")
return
query_record_ids = np.array([x[0] for x in diff_history], dtype=int)
inclusions = np.array([x[1] for x in diff_history], dtype=int)
query_idx = convert_id_to_idx(as_data, query_record_ids)
# Classify the new labels, train and store the results.
with open_state(state_file) as state:
reviewer.classify(
query_idx, inclusions, state, method=label_method)
reviewer.train()
reviewer.log_probabilities(state)
new_query_idx = reviewer.query(reviewer.n_pool()).tolist()
reviewer.log_current_query(state)
# write the proba to a pandas dataframe with record_ids as index
proba = pd.DataFrame(
{"proba": state.pred_proba.tolist()},
index=pd.Index(as_data.record_ids, name="record_id")
)
# update the pool and output the proba's
# important: pool is sorted on query
with SQLiteLock(
lock_file,
blocking=True,
lock_name="active",
project_id=project_id) as lock:
# read the pool
current_pool = read_pool(project_id)
# diff pool and new_query_ind
current_pool_idx = convert_id_to_idx(as_data, current_pool)
current_pool_idx = frozenset(current_pool_idx)
new_pool_idx = [x for x in new_query_idx if x in current_pool_idx]
# convert new_pool_idx back to record_ids
new_pool = convert_idx_to_id(as_data, new_pool_idx)
# write the pool and proba
write_pool(project_id, new_pool)
write_proba(project_id, proba)
