microsoft/DeepSpeed

Pull Requests Commits

add moe skip and tear down dist pg per test

jeffra committed 4 years ago

51518b4a

turn back on test_configurable_parallel

jeffra committed 4 years ago

8afcbf92

make sure test_path is not pytest triggered

jeffra committed 4 years ago

8d59757d

fix megatron regression, add moe unit test, fix moe ckpt comparison

jeffra committed 4 years ago

b01c94fc

fix for moe and revert port scanning

jeffra committed 4 years ago

27003d3c

skip test_configurable_parallel for now

jeffra committed 4 years ago

352a8813

Merge branch 'staging-amd-v2' of github.com:microsoft/DeepSpeed into staging-amd-v2

jeffra committed 4 years ago

ba3822ba

find open port for unit tests

jeffra committed 4 years ago

e9930391

jeffra committed 4 years ago

a346271b

factor out relative path for unit test files

jeffra committed 4 years ago

533bce07

use relative paths for megatron jsons

jeffra committed 4 years ago

bf029e53

use single req.txt in install.sh

jeffra committed 4 years ago

Verified f0b138d1

fix unit test failure when merging PP ckpt files (#1359)

eltonzheng committed 4 years ago

33860bc0

remove rocm skips and install 1bit on torch12

jeffra committed 4 years ago

376e16d2

run unit tests with installed DS + fix bug in SA tests

jeffra committed 4 years ago

562fbb8a

jeffra committed 4 years ago

f4b360ab

[squash] staging amd progress (#1353)

jeffra committed 4 years ago

d4056447

Added 4-byte alignment on NCCL/RCCL (#1328)

amathews-amd committed 4 years ago

Verified 2c628439

[actions] revert unit-test build name

jeffra committed 4 years ago

Verified 3e7d06ad

[actions] add torch version runner label

jeffra committed 4 years ago

Verified 0a32c3e2

[actions] add master to formatting trigger

jeffra committed 4 years ago

Verified 86b948f9

[actions] update branch triggers

jeffra committed 4 years ago

Verified 8f299be5

[actions] split formatting and unit tests into two jobs (#1339)

jeffra committed 4 years ago

Verified 600db091

Update main.yml (#1338)

jeffra committed 4 years ago

Verified 74f058b9

Use mpu in zero.Init() (#1325)

tjruwase committed 4 years ago

Verified e08c239c

Support client lr schedulers that are not subclass of torch _LRScheduler (#1337)

tjruwase committed 4 years ago

Verified 8e301b6e

Added drop_last to DeepSpeedDataLoader (#1321)

hxri committed 4 years ago

Verified c0b27fb0

Remove duplicate clip grad function in deepspeed (#1333)

awan-10 committed 4 years ago

Verified ddffbae0

update for cuda-11.4 (#1329)

stas00 committed 4 years ago

Verified b9ece257

Use clone to avoid checkpoint bloat (#1326)

tjruwase committed 4 years ago

Verified 336dd089

Older