ChainerCV: a Library for Deep Learning in Computer Vision
This release fixes a bug found in Faster R-CNN model (#266). No new feature is added.
This is the last version with new features for Chainer v1. The v0.4* will be updated when there is a critical bug that needs to be fixed. For users using Chainer v2, please use versions v0.5.0 and higher.
Important Axis-dependent values follow row-column order instead of column-row order.
(y_min, x_min, y_max, x_max)
instead of (x_min, y_min, x_max, y_max)
. (#246)(y, x)
. (#235)x_flip, y_flip
for transforms/image
are swapped. (#236)Other changes
eval_semantic_segmentation
. They are now eval_semantic_segmentation_iou
(#215, #217).eval_detection_voc
to eval_detection_voc_ap
. (#232)add_apply_prediction_to_iterator
is added. (#229)generate_random_bbox
(#219)DetectionVocEvaluator
can now report AP for each class. (#242)DetectionVocEvaluator
inside the train code for Faster R-CNN. (#213)eval_voc07
(#231)test_faster_rcnn_train_chain
(#212)eval_voc_07
(#239)