Skip to content

Fix uniform negative sampling #4

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
Edward-Sun merged 1 commit into from
Mar 10, 2019
Merged

Fix uniform negative sampling #4

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
8 changes: 4 additions & 4 deletions codes/model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,7 @@ def forward(self, sample, mode='single'):
In the 'head-batch' or 'tail-batch' mode, sample consists two part.
The first part is usually the positive sample.
And the second part is the entities in the negative samples.
Becuase negative samples and positive samples usually share two elements
Because negative samples and positive samples usually share two elements
in their triple ((head, relation) or (relation, tail)).
'''

Expand Down Expand Up @@ -267,7 +267,7 @@ def train_step(model, optimizer, train_iterator, args):
negative_score = model((positive_sample, negative_sample), mode=mode)

if args.negative_adversarial_sampling:
#In self-negative sampling, we do not apply back-propagation on the sampling weight
#In self-adversarial sampling, we do not apply back-propagation on the sampling weight
negative_score = (F.softmax(negative_score * args.adversarial_temperature, dim = 1).detach()
* F.logsigmoid(-negative_score)).sum(dim = 1)
else:
Expand All @@ -278,8 +278,8 @@ def train_step(model, optimizer, train_iterator, args):
positive_score = F.logsigmoid(positive_score).squeeze(dim = 1)

if args.uni_weight:
positive_sample_loss = positive_score.mean()
negative_sample_loss = negative_score.mean()
positive_sample_loss = - positive_score.mean()
negative_sample_loss = - negative_score.mean()
else:
positive_sample_loss = - (subsampling_weight * positive_score).sum()/subsampling_weight.sum()
negative_sample_loss = - (subsampling_weight * negative_score).sum()/subsampling_weight.sum()
Expand Down