Skip to content

Enable the SSD loss to support normalization by the total number of output locations. #8630

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Feb 28, 2018
Merged

Enable the SSD loss to support normalization by the total number of output locations. #8630

Changes from 3 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
2591fc7
Register more data type for reshape operator.
qingqing01 Feb 27, 2018
a7a3c4f
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
qingqing01 Feb 28, 2018
c6304cb
Enable the SSD loss to support normalization by the total number of o…
qingqing01 Feb 28, 2018
bc4129c
Fix the doc format.
qingqing01 Feb 28, 2018
c7c8d93
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
qingqing01 Feb 28, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 12 additions & 2 deletions python/paddle/fluid/layers/detection.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -328,6 +328,7 @@ def ssd_loss(location,
conf_loss_weight=1.0,
match_type='per_prediction',
mining_type='max_negative',
normalize=True,
sample_size=None):
"""
**Multi-box loss layer for object dection algorithm of SSD**
Expand Down Expand Up @@ -380,14 +381,16 @@ def ssd_loss(location,
neg_overlap (float): The negative overlap upper bound for the unmatched
predictions. Use only when mining_type is max_negative,
0.5 by default.
sample_size (int): The max sample size of negative box, used only when
mining_type is hard_example.
loc_loss_weight (float): Weight for localization loss, 1.0 by default.
conf_loss_weight (float): Weight for confidence loss, 1.0 by default.
match_type (str): The type of matching method during training, should
be 'bipartite' or 'per_prediction', 'per_prediction' by defalut.
mining_type (str): The hard example mining type, should be 'hard_example'
or 'max_negative', now only support `max_negative`.
normalize (bool): Whether to normalize the SSD loss by the total number
of output locations, True by defalut.
sample_size (int): The max sample size of negative box, used only when
mining_type is hard_example.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

hard_example --> 'hard_example'

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Done. Thanks!


Returns:
Variable: The weighted sum of the localization loss and confidence loss,
Expand Down Expand Up @@ -507,6 +510,13 @@ def __reshape_to_2d(var):

# 5.3 Compute overall weighted loss.
loss = conf_loss_weight * conf_loss + loc_loss_weight * loc_loss
# reshape to [N, Np], N is the batch size and Np is the prior box number.
loss = ops.reshape(x=loss, shape=[-1, num_prior])
loss = nn.reduce_sum(loss, dim=1, keep_dim=True)
if normalize:
normalizer = nn.reduce_sum(target_loc_weight)
loss = loss / normalizer

return loss


Expand Down