Bump deepspeed from 0.6.5 to 0.15.1 in /EVA-CLIP

[dependabot[bot]]

Bumps deepspeed from 0.6.5 to 0.15.1.

Release notes

Sourced from deepspeed's releases.

v0.15.1 Patch release

What's Changed

• Update version.txt after 0.15.0 release by @​loadams in microsoft/DeepSpeed#6403
• Fix Type Mismatch by @​jomayeri in microsoft/DeepSpeed#6410
• Fix redundant seq data parallel grp argument in Z3/MiCS by @​samadejacobs in microsoft/DeepSpeed#5352
• add Huawei Ascend NPU setup guide by @​xuedinge233 in microsoft/DeepSpeed#6445
• Add documentation for launcher without SSH by @​dogacancolak-kensho in microsoft/DeepSpeed#6455
• Dtype support check for accelerator in UTs by @​raza-sikander in microsoft/DeepSpeed#6360
• Store/Load CIFAR from local/offline by @​raza-sikander in microsoft/DeepSpeed#6390
• Add the accelerator setup guide link in Getting Started page by @​rogerxfeng8 in microsoft/DeepSpeed#6452
• Allow triton==3.0.x for fp_quantizer by @​siddartha-RE in microsoft/DeepSpeed#6447
• Change GDS to 1 AIO thread by @​jomayeri in microsoft/DeepSpeed#6459
• [CCL] fix condition issue in ccl.py by @​YizhouZ in microsoft/DeepSpeed#6443
• Avoid gds build errors on ROCm by @​rraminen in microsoft/DeepSpeed#6456
• TestLowCpuMemUsage UT get device by device_name by @​raza-sikander in microsoft/DeepSpeed#6397
• Add workflow to build DS without torch to better test before releases by @​loadams in microsoft/DeepSpeed#6450
• Fix patch for parameter partitioning in zero.Init() by @​tohtana in microsoft/DeepSpeed#6388
• Add default value to "checkpoint_folder" in "load_state_dict" of bf16_optimizer by @​ljcc0930 in microsoft/DeepSpeed#6446
• DeepNVMe tutorial by @​tjruwase in microsoft/DeepSpeed#6449
• bf16_optimizer: fixes to different grad acc dtype by @​nelyahu in microsoft/DeepSpeed#6485
• print warning if actual triton cache dir is on NFS, not just for default by @​jrandall in microsoft/DeepSpeed#6487
• DS_BUILD_OPS should build only compatible ops by @​tjruwase in microsoft/DeepSpeed#6489
• Safe usage of popen by @​tjruwase in microsoft/DeepSpeed#6490
• Handle an edge case where CUDA_HOME is not defined on ROCm systems by @​amorehead in microsoft/DeepSpeed#6488

New Contributors

• @​xuedinge233 made their first contribution in microsoft/DeepSpeed#6445
• @​siddartha-RE made their first contribution in microsoft/DeepSpeed#6447
• @​ljcc0930 made their first contribution in microsoft/DeepSpeed#6446
• @​jrandall made their first contribution in microsoft/DeepSpeed#6487
• @​amorehead made their first contribution in microsoft/DeepSpeed#6488

Full Changelog: https://github.com/microsoft/DeepSpeed/compare/v0.15.0...v0.15.1

DeepSpeed v0.15.0

What's Changed

• Update version.txt after 0.14.5 release by @​loadams in microsoft/DeepSpeed#5982
• move pynvml install to setup.py by @​Rohan138 in microsoft/DeepSpeed#5840
• add moe topk(k>2) gate support by @​inkcherry in microsoft/DeepSpeed#5881
• Move inf_or_nan_tracker to cpu for cpu offload by @​BacharL in microsoft/DeepSpeed#5826
• Enable dynamic shapes for pipeline parallel engine inputs by @​tohtana in microsoft/DeepSpeed#5481
• Add and Remove ZeRO 3 Hooks by @​jomayeri in microsoft/DeepSpeed#5658
• DeepNVMe GDS by @​jomayeri in microsoft/DeepSpeed#5852
• Pin transformers version on nv-nightly by @​loadams in microsoft/DeepSpeed#6002
• DeepSpeed on Window blog by @​tjruwase in microsoft/DeepSpeed#6364
• Bug Fix 5880 by @​jomayeri in microsoft/DeepSpeed#6378
• Update linear.py compatible with torch 2.4.0 by @​terry-for-github in microsoft/DeepSpeed#5811
• GDS Swapping Fix by @​jomayeri in microsoft/DeepSpeed#6386
• Long sequence parallelism (Ulysses) integration with HuggingFace by @​samadejacobs in microsoft/DeepSpeed#5774
• reduce cpu host overhead when using moe by @​ranzhejiang in microsoft/DeepSpeed#5578

... (truncated)

Commits

• 10ba3dd Handle an edge case where CUDA_HOME is not defined on ROCm systems (#6488)
• 662a421 Safe usage of popen (#6490)
• ddd3571 Add default value to "checkpoint_folder" in "load_state_dict" of bf16_optimiz...
• 5d1a30c DS_BUILD_OPS should build only compatible ops (#6489)
• ddeb0c1 Fix patch for parameter partitioning in zero.Init() (#6388)
• 9d17116 print warning if actual triton cache dir is on NFS, not just for default (#6487)
• 5df12a4 DeepNVMe tutorial (#6449)
• cfc6ed3 bf16_optimizer: fixes to different grad acc dtype (#6485)
• 9b7fc54 Add workflow to build DS without torch to better test before releases (#6450)
• 89c4d9f TestLowCpuMemUsage UT get device by device_name (#6397)
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot merge` will merge this PR after your CI passes on it - `@dependabot squash and merge` will squash and merge this PR after your CI passes on it - `@dependabot cancel merge` will cancel a previously requested merge and block automerging - `@dependabot reopen` will reopen this PR if it is closed - `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually - `@dependabot show ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/microsoft/LLM2CLIP/network/alerts).