* Add non-interactive prompt, causing issues for some users

* Update pytorch version too

Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>

Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>
Co-authored-by: NJeff Rasley <jerasley@microsoft.com>

Co-authored-by: NJeff Rasley <jerasley@microsoft.com>
Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>

* Fix typo in name of hybrid engine function

* Fix

* Remove dead code

params_already_reduced is not used

* Prevent evaluation of debug strings

Debug strings are evaluated even when logging is disabled

* Use contiguous gradients tensor reduce scatter between ranks

Use allreduce instead of reduce scatter. lower cpu overhead.

* move overflow tracker to optimizer.step

Don't check overflow in gradients for every bucket.
Do overflow chack once on grad flat buffer just before optimizer step

---------
Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>

* DeepSpeed overview in Japanese

* DeepSpeed overview in Japanese

* Small tweak on cuda version mismatch documentation

* clarify minor versions should also match

---------
Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>

* mix typo and missing epsilon value

* Touch file to re-build

* revert changes

* Touch file to re-build

* Format

---------
Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>
Co-authored-by: NLogan Adams <loadams@microsoft.com>

Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>

* fix gpt-j inference issue for mlp_gemm_func call

* bring back the gpt-j inference-test

* fix formatting

* fix the neox and pythia injection issue

This reverts commit f2f5f21b.

Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>
Co-authored-by: NMichael Wyatt <michaelwyatt@microsoft.com>

Co-authored-by: NMichael Wyatt <michaelwyatt@microsoft.com>

Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>

Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>
Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>

Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>
Co-authored-by: NMichael Wyatt <michaelwyatt@microsoft.com>
Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>

* Use logger in accelerator

* Handle pre-build cases

* Explain possible import failure

* fix typo deepspeed/runtime

* fix some typo

---------
Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>

* fix mics save checkpoint hanging

* MiCS load_checkpoint

* copyright

* fix for torch-1.9.0

all_reduce_coalesced api does not support nccl backend

* Naming alignment

* adding more test conditions for mics shard size

* test with different shard sizes

* adding assertion for better error msg

---------
Co-authored-by: NZhen Zhang <zhzhn@amazon.com>

* Refactor check_enabled root validator in DeepSpeedMonitorConfig

* formatting

* formatting

---------
Co-authored-by: NMichael Wyatt <michaelwyatt@microsoft.com>
Co-authored-by: NMichael Wyatt <mrwyattii@gmail.com>

Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>

When activation checkpointing is enabled, most of forward is re-computed,
and so the FLOPS calculation should be updated with recompute_fwd_factor=1.0

I don't find a way to pass the option from model script to deepspeed engine,
and so add option directly for flops_profiler.
Co-authored-by: NCheng Li <pistasable@gmail.com>

* fix spelling error with deepspeed/runtime/

* fix typo docs/

* fix typo in comments with deepspeed/

* fix typo deepspeed/

* Update constants.py

Remove the space after nebula

---------
Co-authored-by: NLogan Adams <114770087+loadams@users.noreply.github.com>
Co-authored-by: NMichael Wyatt <michaelwyatt@microsoft.com>

Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>

Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>

Co-authored-by: NMichael Wyatt <michaelwyatt@microsoft.com>

Code (in this context) is mass noun, and thus has no plural form.
Co-authored-by: NMichael Wyatt <michaelwyatt@microsoft.com>

grammar fix.
Co-authored-by: NMichael Wyatt <michaelwyatt@microsoft.com>

* skip test for docs-only changes

* add missing skip to blog changes

* add Ascend NPU accelerator support

* clean code

---------
Co-authored-by: Njializheng <jializheng@huawei.com>
Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>

this change also aligns with the logic before reduce_scatter_coalesced
Co-authored-by: NOlatunji Ruwase <olruwase@microsoft.com>

* tmp remove launcher args

* add exclude list for env variables on aisc

* add comment

* Align InferenceEngine to store ms in _model_times

   When using cuda_events, the measured model time is stored in ms.
   When not using cuda_events, the measured model time was stored in seconds.
   This commit fixes the units and aligns them to store ms, the same as elapsed() function.
   This was observed when running the following pytest:
   unit/inference/test_model_profiling.py::TestModelProfiling::test[False-True-roberta-base-fill-mask]

   Returned values were:
     count=0 e2e_t=895.174312 model_t=0.8529715538024902
     count=1 e2e_t=7.500252 model_t=0.0041310787200927734
     count=2 e2e_t=3.887346 model_t=0.0018568038940429688
     count=3 e2e_t=3.577845 model_t=0.0016334056854248047
     count=4 e2e_t=3.43976 model_t=0.0016703605651855469
     count=5 e2e_t=3.310903 model_t=0.0016107559204101562
     count=6 e2e_t=3.299556 model_t=0.001603841781616211
     count=7 e2e_t=3.605722 model_t=0.0015969276428222656
     count=8 e2e_t=3.273741 model_t=0.0015516281127929688
     count=9 e2e_t=3.46306 model_t=0.0016617774963378906

   The units difference is observed here, when model_t is in ther order of 10e-3 comparing to e2e_t

* Update engine.py

---------
Co-authored-by: NMichael Wyatt <michaelwyatt@microsoft.com>

Co-authored-by: NJeff Rasley <jerasley@microsoft.com>