Skip to content
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

FIX: fix the loss weight of S3Rec #2028

Merged
merged 3 commits into from
Mar 30, 2024
Merged

FIX: fix the loss weight of S3Rec #2028

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
ae720aa
Fix: fix the loss coefficient of s3rec.
zhengbw0324 Mar 19, 2024
d6567a8
Merge branch 'RUCAIBox:master' into master
zhengbw0324 Mar 19, 2024
502e25c
Merge branch 'RUCAIBox:master' into master
zhengbw0324 Mar 30, 2024
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
6 changes: 3 additions & 3 deletions recbole/properties/model/S3Rec.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,12 +9,12 @@ layer_norm_eps: 1e-12 # (float) A value added to the denominator for n
initializer_range: 0.02 # (float) The standard deviation for normal initialization.
item_attribute: 'class' # (str) The item features used as attributes for pre-training.
mask_ratio: 0.2 # (float) The probability for a item replaced by MASK token.
aap_weight: 1.0 # (float) The weight for Associated Attribute Prediction loss.
mip_weight: 0.2 # (float) The weight for Masked Item Prediction loss.
aap_weight: 0.2 # (float) The weight for Associated Attribute Prediction loss.
mip_weight: 1.0 # (float) The weight for Masked Item Prediction loss.
map_weight: 1.0 # (float) The weight for Masked Attribute Prediction loss.
sp_weight: 0.5 # (float) The weight for Segment Prediction loss.
train_stage: 'pretrain' # (str) The training stage. Range in ['pretrain', 'finetune'].
pretrain_epochs: 500 # (int) The epochs of pre-training.
save_step: 10 # (int) Save pre-trained model every pre-training epochs.
save_step: 50 # (int) Save pre-trained model every pre-training epochs.
pre_model_path: '' # (str) The path of pretrained model.
loss_type: 'CE' # (str) The type of loss function. Range in ['BPR', 'CE'].
Loading