DeepSpeed
add SD injection policy
#2381
Merged

add SD injection policy #2381

jeffra merged 29 commits into master from jeffra/sd-policy
jeffra
jeffra add initial sd policy
df895f7a
jeffra Merge branch 'master' into jeffra/sd-policy
d0576d48
jeffra formatting
c8163e28
add attention kernel and enable cuda-graph for SD models
f34eb5bb
add new files
5aec40c2
jeffra formatting and add dtype to unet
23674fd8
adding more optitmization by enabling ds-encoder with CUDA-Graph
6bf0c738
RezaYazdaniAminabadi Merge branch 'master' into jeffra/sd-policy
8b0762b9
add missing file
ac29fdcc
Merge branch 'jeffra/sd-policy' of github.com:microsoft/DeepSpeed int…
c1770708
adapt the triton kernel to be used in more places
27e752be
add more fusion
a40780b5
allocate workspace using the padded hidden_size
923ddd2f
skip the clip-encoder injection for now
2dca3acd
jeffra Merge branch 'master' into jeffra/sd-policy
52dd412e
jeffra add triton to new extra
6079065f
jeffra lazy import triton, add sd extra, formatting
fb1605f3
jeffra jeffra marked this pull request as ready for review 3 years ago
jeffra jeffra requested a review from samyam samyam 3 years ago
jeffra jeffra requested a review from tjruwase tjruwase 3 years ago
jeffra jeffra requested a review from ShadenSmith ShadenSmith 3 years ago
jeffra jeffra requested a review from conglongli conglongli 3 years ago
jeffra jeffra requested a review from awan-10 awan-10 3 years ago
jeffra jeffra requested a review from cli99 cli99 3 years ago
jeffra jeffra requested a review from eltonzheng eltonzheng 3 years ago
jeffra jeffra requested a review from minjiaz minjiaz 3 years ago
jeffra jeffra requested a review from RezaYazdaniAminabadi RezaYazdaniAminabadi 3 years ago
jeffra jeffra requested a review from duli2012 duli2012 3 years ago
jeffra jeffra requested a review from mrwyattii mrwyattii 3 years ago
jeffra jeffra requested a review from yaozhewei yaozhewei 3 years ago
jeffra jeffra requested a review from arashb arashb 3 years ago
jeffra jeffra requested a review from xiaoxiawu-microsoft xiaoxiawu-microsoft 3 years ago
jeffra jeffra requested a review from samadejacobs samadejacobs 3 years ago
jeffra jeffra requested a review from cmikeh2 cmikeh2 3 years ago
jeffra jeffra requested a review from GuanhuaWang GuanhuaWang 3 years ago
jeffra delay import
16bcef13
jeffra fix previous issue i added
691fd349
fix bug with adding bias
c758b03e
RezaYazdaniAminabadi Merge branch 'master' into jeffra/sd-policy
da013373
jeffra fixes for triton import and add acks to triton-ops file
406832c5
jeffra Merge branch 'jeffra/sd-policy' of github.com:microsoft/DeepSpeed int…
79b05ca1
jeffra Merge branch 'master' into jeffra/sd-policy
a36275c1
merge fix & formatting
75fbcfe7
Merge branch 'jeffra/sd-policy' of github.com:microsoft/DeepSpeed int…
e6175955
fix small issue
eff95e7e
skip cuda-graph for clip-encoder for now (it has issue on larger batc…
770e88b1
jeffra Merge branch 'master' into jeffra/sd-policy
31488815
jeffra jeffra merged ec13da6b into master 3 years ago
jeffra jeffra deleted the jeffra/sd-policy branch 3 years ago

Login to write a write a comment.

Login via GitHub