Summary: 45 instances, 40 unique

Text	Count
# TODO: in the case where we're loading an existing set of FPs after manual filtering,	1
# TODO: add the petrained weights url map of 'efficientnet-l2'	1
print('runserver, post_detect_sync, user specified detection confidence: ', detection_confidence)  # TODO delete	1
// TODO: confirm json file is reading in detections correctly	1
# TODO: convert to a blocking read and eliminate the sleep() statement in this loop	1
#  TODO update: Assumes that crops have already	1
# TODO 'classification mean inference time': str(''),	1
// display the images and the current predictions; TODO: should not color based on detection type 0 (ActiveDetection)--these are actually images flagged for labeling but may also have a machine predicted label	2
# TODO: change weighted to False if oversampling minority classes	1
# TODO hardcoded columns: change if # of examples or col_order changes	1
TODO: potential speedup by avoiding PNG compression and PIL dependency	1
# TODO: iterrows() is a terrible way to do this, but this is one of those days	2
# TODO log exception when we have more telemetry	1
# TODO do not read the entry first to get the call_params when the Cosmos SDK add a	1
# TODO classifier = TFClassifier(api_config.CLASSIFICATION_MODEL_PATHS, api_config.CLASSIFICATION_CLASS_NAMES)	1
TODO - Copy this module from the classification folder once it has been refactored so that we do not keep	1
# TODO check that the expiry date of input_container_sas is at least a month	1
# TODO: hit some silly issues with vectorized str() and escaped characters, vectorize	1
# TODO move detector_output_path specific code out so that this function evaluates only on classification results (confidences)	1
// TODO: change this code copied from web	1
# TODO: consider weighting val and test set as well	1
dataset_query = Detection.select(Detection.image_id, Oracle.label, Detection.kind).join(Oracle).order_by(fn.random()).limit(args.db_query_limit) ## TODO: should this really be order_by random?	1
# TODO: should this also change self.kind?	1
temp_labels[j] = True #TODO: this currently only works for oneclass?	1
s = img.shape; h = s[0]; w = s[1]  #TODO: check this isn't backwards	1
# TODO: read from memory rather than using intermediate files	1
temp_labels[j] = True #TODO: this currently only works for oneclass?	2
# TODO - check based on access policy as well	1
'location': location  # TODO bug - location should be region_name + camera_location	1
// TODO: add error checking return null on error	1
#    temp_labels[j] = True #TODO: this currently only works for oneclass?	1
for i_row, cell in enumerate(ws['D']):  # TODO hardcoded column number	1
# TODO: convert to a blocking read and eliminate the sleep() statement in this loop	1
args.strategy = 'confidence' ## TODO: hard coding this for now	2
# TODO: optionally write html only for images where rendering succeeded	1
if model_version not in api_config.MD_VERSIONS_TO_REL_PATH:  # TODO use AppConfig to store model version info	1
## TODO: allow user to update the class list through the labeling tool UI as they see different species	2
# TODO: modify the params names.	1
Possible TODO: oversample the imbalanced classes	1
# TODO add MD versions info to AppConfig	1