issues
search
facebookresearch
/
generative-recommenders
Repository hosting code used to reproduce results in "Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations" (https://arxiv.org/abs/2402.17152, ICML'24).
Apache License 2.0
466
stars
77
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Set seq_len and n_targets int32
#50
xing-liu
opened
7 hours ago
3
Removes INVALID_MASK_TYPE == "none_lower_triangular"
#49
xing-liu
opened
8 hours ago
1
reformats codebase.
#48
jiaqizhai
closed
17 hours ago
1
Fix debug config for attention kernels
#47
xing-liu
opened
1 day ago
1
Question about the usage of torch.ops.fbgemm.dense_to_jagged and torch.ops.fbgemm.jagged_to_padded_dense
#46
KkEeVvIiNnn
closed
1 day ago
1
is rating not used?
#45
yueming-zhang
closed
1 day ago
1
cleans up requirements.txt per automated checks
#44
jiaqizhai
opened
6 days ago
1
adding bias bwd makes fused attention bwd 50x slower, any suggestions?
#43
zzq96
opened
6 days ago
1
Are we using oldest rating as target?
#42
yundai424
closed
1 week ago
2
_ragged_hstu_attn_fwd triton offset optimizations
#41
zhaozhul
closed
1 week ago
1
Adds example requirements.txt
#40
jiaqizhai
closed
1 week ago
2
Question of Behaviour Sequence Length on Public Dataset
#39
djx2726889
opened
1 week ago
1
Use input_precision="tf32x3" when allow_tf32=true
#38
htyu
opened
2 weeks ago
2
move group_norm kernel to generative_recommenders
#37
LinjianMa
closed
2 weeks ago
2
Regarding padding, gr_output_length
#36
wookjeHan
closed
1 week ago
1
CUDA and PyTorch Versions
#35
rtalaricw
opened
2 weeks ago
3
Optimize autotune configs for `_addmm_fwd` kernel
#34
sijiac
closed
1 week ago
4
triton flash attention with bias backward.
#33
NAP-GHJ
opened
2 weeks ago
5
Move addmm_fwd to generative_recommenders folder
#32
cx-yin
closed
2 weeks ago
5
Stochastic Length strategy may cause user history features inconsistent online and offline
#31
muGithub
opened
3 weeks ago
0
How to combine autoregressive model and classification multi-task head in ranking model?
#30
MoFHeka
opened
3 weeks ago
1
add a unittest for ragged attention
#29
LinjianMa
closed
3 weeks ago
1
The detail about the generation of tokens?
#28
glb400
opened
4 weeks ago
3
Triton code
#27
wookjeHan
closed
2 weeks ago
1
GRs use yapf to format code
#26
Mr-Nineteen
closed
4 weeks ago
1
The code is not formatted.
#25
Mr-Nineteen
closed
17 hours ago
2
RuntimeError: CUDA error: invalid device ordinal
#24
china10s
closed
1 month ago
1
Add position index input to support different timestamp chunks in HSTU attwntion kernel for context parallelism
#23
yoyoyocmu
opened
1 month ago
4
Separate normalization factor from max_seq_len
#22
yoyoyocmu
opened
1 month ago
3
Support no-mask in the HSTU attention kernel in both forward and backward
#21
yoyoyocmu
opened
1 month ago
3
Update ragged_hstu_attn autotune cache
#20
cx-yin
closed
1 month ago
1
token的组织形式
#19
llmmodel123
closed
1 day ago
1
AttributeError: '_OpNamespace' 'fbgemm' object has no attribute 'jagged_2d_to_dense'
#18
ShuaiXuan
closed
1 month ago
4
Support no-mask in the HSTU attention kernel in both forward and backward
#17
yoyoyocmu
closed
1 month ago
2
add copyright to oss triton kernel
#16
LinjianMa
closed
1 month ago
2
Re-sync with internal repository
#15
facebook-github-bot
closed
1 month ago
0
question about RelativeBucketedTimeAndPositionBasedBias
#14
ShuaiXuan
closed
1 month ago
1
updates README.md
#13
jiaqizhai
closed
1 month ago
2
question about the action part
#12
levi77
closed
1 month ago
1
fix: fix high-probability directory creation race condition error
#11
kitstar
closed
1 month ago
4
removes deprecated item_feature_embedding_dim
#10
jiaqizhai
closed
1 month ago
0
perhaps remove item_feature_embedding_dim in gin config for amzn-books?
#9
zhuochenKIDD
closed
2 months ago
1
Possible discover a dataset bug.
#8
cb521
closed
2 months ago
1
Run error
#7
cb521
closed
2 months ago
2
rank demo
#6
a6802739
closed
2 months ago
1
user-item interaction question
#5
a6802739
closed
1 month ago
1
run error
#4
a6802739
closed
2 months ago
3
code cleanup 2/2.
#3
jiaqizhai
closed
2 months ago
3
Stochastic Length
#2
djx2726889
closed
1 month ago
1
TypeError: __call__() got an unexpected keyword argument 'target_ids'
#1
daiwk
closed
2 months ago
3
Next