DeepSpeed
Parallel map step for `DistributedDataAnalyzer` map-reduce
#5291
Merged

Parallel map step for `DistributedDataAnalyzer` map-reduce #5291

conglongli merged 60 commits into deepspeedai:master from parallel_run_in_distributed_data_analyzer
bm-synth
bm-synth added assert of torch vs numpy types
14f2bbe1
bm-synth first draft
796341d9
bm-synth reverted to original master
07aa4b42
bm-synth added metric type accumulate_value_over_samples
815a7897
bm-synth pre-commit
28a72e7c
bm-synth Merge branch 'master' into distributed_data_analyzer
e8dbf0b3
bm-synth Merge branch 'distributed_data_analyzer' of github.com:bm-synth/DeepS…
ec3479fa
bm-synth Update data_analyzer.py
38d7ce66
bm-synth added check for single node reduce. added barriers
295fba67
bm-synth more bug fixes
4144e427
bm-synth new iteration, many bug fixes
a1e121c9
bm-synth bug fixes
e045753c
bm-synth Merge branch 'master' into distributed_data_analyzer
3a891162
bm-synth fixing previous commit
cdc838c1
bm-synth Merge branch 'master' into distributed_data_analyzer
ba34a550
bm-synth pre-commit
5c077104
bm-synth Merge branch 'distributed_data_analyzer' of github.com:bm-synth/DeepS…
87d76867
bm-synth recoverd master branch
f28e829b
bm-synth write sequentially to file
a634787f
bm-synth Merge branch 'master' into distributed_data_analyzer
848ffd5d
bm-synth fixes in sequential write
ec59f08d
bm-synth Merge branch 'distributed_data_analyzer' of github.com:bm-synth/DeepS…
832874c2
bm-synth pre-commit hooks
ea0d65f5
bm-synth Merge branch 'master' into distributed_data_analyzer
c6c9bc5b
bm-synth added main as example
56a95338
bm-synth Merge branch 'distributed_data_analyzer' of github.com:bm-synth/DeepS…
b4d86543
bm-synth Merge branch 'master' into distributed_data_analyzer
676dc1a3
bm-synth Update data_analyzer.py
6788af55
bm-synth first working version. idx files differ
bd61d9c2
bm-synth Merge branch 'distributed_data_analyzer' of github.com:bm-synth/DeepS…
7ac5e45c
bm-synth added missing static function
8bf0e635
bm-synth removed/added breaklines to match base code
e5a7eb0f
bm-synth corrected comment
3b8014fd
bm-synth imports
5a426879
bm-synth removed main
cdaad362
bm-synth reverted main
b3d40620
bm-synth bug fix in sample calculation
7cabfa2a
bm-synth added worker_an and num_worker to kwargs
62f68dd1
bm-synth removed dist.initialize ()from DataAnalyzer.run_map_reduce
6d35e454
bm-synth first iteration
be91d37c
bm-synth updated with add_items
5fd05468
bm-synth Merge branch 'master' into serial_data_analyzer
e943aaa7
bm-synth master in line with remote
f6c5c18d
bm-synth merge with master
2e2ebee4
bm-synth first iteration, needs testing and debugging on num_threads>1
e4de5545
bm-synth first draft of multiprocessing
dadf0f92
bm-synth first draft of multiprocessing
0ba9f930
bm-synth pre-commit hooks
a761196a
bm-synth bm-synth changed the title Parallel run in distributed data analyzer Parallel map step for `DistributedDataAnalyzer` map-reduce 1 year ago
bm-synth Merge branch 'master' into parallel_run_in_distributed_data_analyzer
7b9d1c66
bm-synth pre-commit hooks and merge conflicts
c5dad209
bm-synth bm-synth marked this pull request as ready for review 1 year ago
bm-synth bm-synth requested a review from conglongli conglongli 1 year ago
bm-synth faster writes
4313e8df
bm-synth pre-commit hooks
34965a04
bm-synth Merge branch 'master' into parallel_run_in_distributed_data_analyzer
fb5f8fa6
bm-synth Merge branch 'master' into parallel_run_in_distributed_data_analyzer
2c30081a
bm-synth Merge branch 'master' into parallel_run_in_distributed_data_analyzer
6d26d1aa
bm-synth Merge branch 'master' into parallel_run_in_distributed_data_analyzer
afa7495d
bm-synth Merge branch 'master' into parallel_run_in_distributed_data_analyzer
6106757e
bm-synth
loadams Merge branch 'master' into parallel_run_in_distributed_data_analyzer
6abfdda5
bm-synth Merge branch 'master' into parallel_run_in_distributed_data_analyzer
444644b2
conglongli Merge branch 'master' into parallel_run_in_distributed_data_analyzer
b130322f
conglongli
conglongli approved these changes on 2024-04-18
conglongli conglongli assigned conglongli conglongli 1 year ago
conglongli conglongli enabled auto-merge 1 year ago
conglongli conglongli merged 64defe65 into master 1 year ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
Labels
Milestone