microsoft/DeepSpeed

Pull Requests Commits

Merge branch 'master' into olruwase/nvme_offload_bug

tjruwase committed 4 years ago

Verified a2c4af4b

bump 0.3.17 for next release

jeffra committed 4 years ago

Verified e4139a7d

Merge branch 'master' into olruwase/nvme_offload_bug

tjruwase committed 4 years ago

Verified 369c02e5

fix assert message (#1040)

stas00 committed 4 years ago

Verified 47ec97eb

Change methods to be static (#1038)

janEbert committed 4 years ago

Verified 21047072

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

tjruwase committed 4 years ago

bb060ee4

Add fix message for missing aio lib error.

tjruwase committed 4 years ago

f892b655

Merge branch 'master' into olruwase/nvme_offload_bug

tjruwase committed 4 years ago

Verified 39d634c9

fix format (#1032)

cli99 committed 4 years ago

Verified 962dbc63

NVMe intra-request validation should be on entire file

tjruwase committed 4 years ago

8d4817ba

[Stage][Fix] Add additional conditions when checking types of output from the model (#1026)

Sean Naren committed 4 years ago

Verified b3870363

Bump rexml from 3.2.4 to 3.2.5 in /docs (#1025)

dependabot[bot] committed 4 years ago

Verified c4f4ef5d

Validate env; remove write size from logs (#1013)

tjruwase committed 4 years ago

Verified 820515e4

jeffra committed 4 years ago

a12aaedc

Remove duplicated _initialize_parameter_parallel_groups definition in engine (#709)

zhujiangang committed 4 years ago

Verified 6da4fccc

[tests] make it easier to run tests (#923)

stas00 committed 4 years ago

Verified de694b91

Relax dataset type check in deepspeed io (#1012)

cli99 committed 4 years ago

Verified c5700bc0

improve assert message (#1024)

stas00 committed 4 years ago

Verified c4485e2f

[fp32] fix default dtype (#1023)

stas00 committed 4 years ago

Verified 18a26e86

Handle Norm allreduce when no mp (#1021)

tjruwase committed 4 years ago

Verified 429dfa6c

Samyamr/full precision for ZeRO Stage2 and Stage3 (#1004)

samyam committed 4 years ago

Verified dad26428

Transformer - fix unit test (#964)

RezaYazdaniAminabadi committed 4 years ago

Verified cab30aa6

[zero_to_fp32.py] support param groups (#1017)

stas00 committed 4 years ago

Verified a8cf887d

Refactor param_dict to config (#1008)

Sean Naren committed 4 years ago

Verified 41ab660b

add ds integrations (#963)

cli99 committed 4 years ago

Verified 14a50c68

fix gradient_clipping default (#656)

stas00 committed 4 years ago

Verified b7f97061

Add find_unused_parameters option to DeepSpeedEngine (#945)

ghosthamlet committed 4 years ago

Verified d0b61f18

Minor usability tweaks/fixes (#1001)

tjruwase committed 4 years ago

Verified 03d24fe0

Use amp autocast in ZeRO3 linear (#990)

tjruwase committed 4 years ago

Verified e88ebbcf

Asynchronous I/O docs (#1000)

tjruwase committed 4 years ago

Verified bff4bc72

Older