2martens/masterthesis
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Finished vanilla SSD evaluation code

Browse Source 
Signed-off-by: Jim Martens <github@2martens.de>
This commit is contained in:
Jim Martens
2019-05-16 14:07:20 +02:00
parent 2a34bfca3f
commit 9d726ebb10
1 changed files with 23 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
src/twomartens/masterthesis/main.py
View File

@ -174,7 +174,11 @@ def _ssd_test(args: argparse.Namespace) -> None:
batch_size = 16
use_dropout = False if args.network == "ssd" else True
output_path = f"{args.output_path}/val/{args.network}/{args.iteration}"
evaluation_path = f"{args.evaluation_path}/{args.network}"
result_file = f"{evaluation_path}/results-{args.iteration}.bin"
label_file = f"{output_path}/labels.bin"
predictions_file = f"{output_path}/predictions.bin"
predictions_per_class_file = f"{output_path}/predictions_class.bin"
# retrieve labels and un-batch them
files = glob.glob(f"{output_path}/*ssd_labels*")
@ -205,9 +209,15 @@ def _ssd_test(args: argparse.Namespace) -> None:
del _predictions
# prepare predictions for further use
with open(predictions_file, "wb") as file:
pickle.dump(predictions, file)
predictions_per_class = evaluate.prepare_predictions(predictions, ssd.N_CLASSES)
del predictions
with open(predictions_per_class_file, "wb") as file:
pickle.dump(predictions_per_class, file)
# compute matches between predictions and ground truth
true_positives, false_positives, \
cum_true_positives, cum_false_positives = evaluate.match_predictions(predictions_per_class,
@ -221,7 +231,19 @@ def _ssd_test(args: argparse.Namespace) -> None:
average_precisions = evaluate.get_mean_average_precisions(cum_precisions, cum_recalls, ssd.N_CLASSES)
mean_average_precision = evaluate.get_mean_average_precision(average_precisions)
# TODO store result of evaluation
results = {
"true_positives": true_positives,
"false_positives": false_positives,
"cumulative_true_positives": cum_true_positives,
"cumulative_false_positives": cum_false_positives,
"cumulative_precisions": cum_precisions,
"cumulative_recalls": cum_recalls,
"average precisions": average_precisions,
"mean_average_precision": mean_average_precision
}
with open(result_file, "wb") as file:
pickle.dump(results, file)
def _val(args: argparse.Namespace) -> None: