DeepSpeed
Transformer kernel release
#242
Merged

Transformer kernel release #242

jeffra merged 35 commits into master from kernel-staging
jeffra
jeffra Transformer kernels (#49)
06076c63
jeffra update DSE
3b1ef351
jeffra remove warning note about 44min tutorial coming soon
c6311b21
RezaYazdaniAminabadi add the transformer tutorial (#50)
c2ca8073
jeffra bump version number to 0.2.0
385acb33
jeffra only all-reduce grads if dp world size > 1
99c8ed49
jeffra Merge branch 'jeffra/staging' of github.com:microsoft/DeepSpeed-inter…
b8c6c198
jeffra revert previous commit, an issue with zero-2 with this change
d7fa8e1c
eltonzheng add transformer kernel API in website (#51)
05b9749c
samyam Bert Tutorial update (#52)
4291d969
jeffra update DSE
6f043ef9
jeffra add master addr/port to local launching
9d79d817
jeffra minor cleanup of transformer tutorial
883de502
jeffra add intial version of bert deep dive post
c5d80e18
jeffra update img paths for staging
a16088a8
jeffra center table
c2cbb986
jeffra update tput images and table
f1951136
jeffra space between figures
5af86b24
jeffra center table
be8f176d
jeffra update image
fa242988
jeffra un-center table
8f4f96b2
jeffra references
41bbe209
jeffra add softmax animation
c81ffb98
jeffra add laynorm gif
a0a4baad
jeffra update gifs
a73b3610
jeffra add a space between gifs
6c9dbdc2
jeffra update image paths
434dc450
eltonzheng add stochastic_mode in API doc (#53)
f3b14e54
jeffra update images
fea90fe2
jeffra tmp img path for staging
02e3048f
RezaYazdaniAminabadi update the tutorials for fine-tuning (#54)
bf5182e1
jeffra update images
195a7acb
jeffra Merge branch 'jeffra/staging' of github.com:microsoft/DeepSpeed-inter…
9668591d
jeffra fix img path for live
4e7ca29f
jeffra Merge branch 'master' into kernel-staging
af560da8
jeffra jeffra merged 734d8991 into master 5 years ago
jeffra jeffra deleted the kernel-staging branch 5 years ago

Login to write a write a comment.

Login via GitHub

Reviewers
No reviews
Assignees
No one assigned
Labels
Milestone