未验证 提交 c3b96395 编写于 作者: K Kaipeng Deng 提交者: GitHub

add batch_size in YOLOv3Loss for backward compatable. (#1469)

* add batch_size in YOLOv3Loss for backward compatable.
上级 f7144154
......@@ -41,8 +41,10 @@ class YOLOv3Loss(object):
__inject__ = ['iou_loss', 'iou_aware_loss']
__shared__ = ['use_fine_grained_loss', 'train_batch_size']
def __init__(self,
def __init__(
self,
train_batch_size=8,
batch_size=8, # stub for backward compatable
ignore_thresh=0.7,
label_smooth=True,
use_fine_grained_loss=False,
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册