microsoft/DeepSpeed

Pull Requests Commits

Merge branch 'master' into olruwase/zero2_offload_rrb_divergence

tjruwase committed 4 years ago

Verified 7030346c

formatting fix for release script

jeffra committed 4 years ago

d1a7a55e

bump to 0.4.4 + add release scripts

jeffra committed 4 years ago

dba0f431

Merge branch 'master' into olruwase/zero2_offload_rrb_divergence

tjruwase committed 4 years ago

Verified 2a6121c1

revert part of #1220 (#1221)

stas00 committed 4 years ago

Verified bc451c09

Merge branch 'master' into olruwase/zero2_offload_rrb_divergence

tjruwase committed 4 years ago

Verified 68406776

[zero3] release tmp memory when consolidating fp16 weights (#1220)

stas00 committed 4 years ago

Verified 2660cc4d

Merge branch 'master' into olruwase/zero2_offload_rrb_divergence

tjruwase committed 4 years ago

Verified c14f60e9

[model weights] zero_to_fp32 multiple improvements (#1181)

stas00 committed 4 years ago

Verified 2a921069

Merge branch 'master' into olruwase/zero2_offload_rrb_divergence

tjruwase committed 4 years ago

Verified 24ab1ec0

enable cpu adam op on powerpc architectures (#1213)

adammoody committed 4 years ago

Verified f65ff908

Merge branch 'olruwase/zero2_offload_rrb_divergence' of github.com:microsoft/DeepSpeed into

tjruwase committed 4 years ago

406d0219

tjruwase committed 4 years ago

5ef4efa4

Merge branch 'master' into olruwase/zero2_offload_rrb_divergence

tjruwase committed 4 years ago

Verified e3e35d67

Remove debug codes

tjruwase committed 4 years ago

c5099998

Remove debug codes

tjruwase committed 4 years ago

079de3c4

tjruwase committed 4 years ago

b4e23e15

json wrong format (#1210)

GeorgeSen committed 4 years ago

Verified 5652072e

improve debug (#1215)

stas00 committed 4 years ago

Verified 5127b2fa

[zero.Init] post_init partitining is to be run only by a child module (#1202)

stas00 committed 4 years ago

Verified 497b741f

[zero3] params_to_reduce isn't always there (#1214)

stas00 committed 4 years ago

Verified 91f58c06

Upgrade bundler to address CVE-2020-36327

jeffra committed 4 years ago

Verified aeea85b4

contiguous gradients should be set to True by default (#1199)

samyam committed 4 years ago

Verified c9fee821

jeffra committed 4 years ago

36126122

clean up logging (#1190)

stas00 committed 4 years ago

Verified a0292398

undo noise (#1191)

stas00 committed 4 years ago

Verified bc019a53

Fix bugs about non-contiguous tensor broadcasting (#1168)

hyunwoongko committed 4 years ago

Verified 429cbc89

introduce debug utils (#1136)

stas00 committed 4 years ago

Verified c0c4ebf1

ZeRO 2+3 memory estimators (#965)

stas00 committed 4 years ago

Verified 0c1802cc

zero_to_fp32: restore persistent buffers (#1146)

stas00 committed 4 years ago

Verified df8b1f88

Older