Deepspeed-mii

What's Changed
* Add community tutorial by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/104
* Add xlm-roberta by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/113
* Use new DS-Inference config API by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/108
* Improvements to VRAM usage when loading HF models by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/105
* Add Bloom model config checks for HF provider by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/107


**Full Changelog**: https://github.com/microsoft/DeepSpeed-MII/compare/v0.0.3...v0.0.4

What's Changed
* fix typo by TahaBinhuraib in https://github.com/microsoft/DeepSpeed-MII/pull/70
* move deserialize logic into wrapper to match diffusers api by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/77
* bump min deepspeed to 0.7.4 by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/82
* docs(deploy): update handled tasks by Thytu in https://github.com/microsoft/DeepSpeed-MII/pull/81
* fix(example): QnA model name by Thytu in https://github.com/microsoft/DeepSpeed-MII/pull/80
* Fix aio config dict in ZeRO text-gen example by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/88
* Support latest Transformers and new cache design by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/69
* use hf_auth_token in mii_config to huggingface provider by l-yohai in https://github.com/microsoft/DeepSpeed-MII/pull/76
* SD tutorial and benchmarks by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/90
* enable cuda graph for diffusers models by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/96

New Contributors
* TahaBinhuraib made their first contribution in https://github.com/microsoft/DeepSpeed-MII/pull/70
* Thytu made their first contribution in https://github.com/microsoft/DeepSpeed-MII/pull/81
* l-yohai made their first contribution in https://github.com/microsoft/DeepSpeed-MII/pull/76

**Full Changelog**: https://github.com/microsoft/DeepSpeed-MII/compare/v0.0.2...v0.0.3

What's Changed
* Add pkg, versioning, formatting, etc. by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/2
* [CI] add formatter job by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/3
* Local deployment support by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/4
* [docs] add basic install instructions by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/5
* [docs] add contribution guide by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/6
* Samyam/generic server client score by samyam in https://github.com/microsoft/DeepSpeed-MII/pull/7
* add cache-dir, detect crash, clean-up by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/8
* Azure local deployment by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/9
* Changing task names to ENUM by samyam in https://github.com/microsoft/DeepSpeed-MII/pull/10
* Samyamr/model registration by samyam in https://github.com/microsoft/DeepSpeed-MII/pull/11
* Samyamr/add features by samyam in https://github.com/microsoft/DeepSpeed-MII/pull/12
* Samyamr/parallelism by samyam in https://github.com/microsoft/DeepSpeed-MII/pull/13
* Samyamr/non grpc deployment test by samyam in https://github.com/microsoft/DeepSpeed-MII/pull/14
* Samyamr/aks examples by samyam in https://github.com/microsoft/DeepSpeed-MII/pull/15
* Samyamr/core documentation by samyam in https://github.com/microsoft/DeepSpeed-MII/pull/17
* Add huggingface models and tasks that use gpt2 and robert model type by cli99 in https://github.com/microsoft/DeepSpeed-MII/pull/16
* refactor model type check and add bert by cli99 in https://github.com/microsoft/DeepSpeed-MII/pull/19
* [docs] readme updates by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/20
* add gpt_neo and gptj support by cli99 in https://github.com/microsoft/DeepSpeed-MII/pull/21
* Add Pydantic for checking configurations by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/28
* [config] add dtype+validation for config and pass to ds-init by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/30
* gpt-neox support by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/22
* Add unit tests for local deployment by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/31
* Bug Fix for FP16 models by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/32
* GPT-NeoX docs by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/33
* Fix for GPT-NeoX model by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/34
* force hf to use pytorch and not use tf by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/36
* Enable zero inference in a separate code path from ds_optimize by cli99 in https://github.com/microsoft/DeepSpeed-MII/pull/24
* Remove non-local deployments for upcoming release by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/37
* Doc and api update by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/39
* Add Bloom Models by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/38
* Multi-batch support for text-generation by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/40
* expose all kwargs for generate by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/42
* AML deployment + major refactor by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/41
* fix hyperlink to paper by kamalkraj in https://github.com/microsoft/DeepSpeed-MII/pull/48
* Enable generation of scripts and configs for AML deployments by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/55
* Fixed bigscience examples by cderinbogaz in https://github.com/microsoft/DeepSpeed-MII/pull/53
* Allow pinning deployments to specific rank(s) by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/56
* various clean-up by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/59
* AML deployment improvements by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/62
* update supported models by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/63
* ds-azure as preview release by samyam in https://github.com/microsoft/DeepSpeed-MII/pull/64
* New landing + plots by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/65
* Landing page changes by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/67
* Fix examples with wrong param names by mrwyattii in https://github.com/microsoft/DeepSpeed-MII/pull/68
* add support for txt2img task by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/57
* use the fp16 revision of SD by jeffra in https://github.com/microsoft/DeepSpeed-MII/pull/72

New Contributors
* cli99 made their first contribution in https://github.com/microsoft/DeepSpeed-MII/pull/16
* kamalkraj made their first contribution in https://github.com/microsoft/DeepSpeed-MII/pull/48
* cderinbogaz made their first contribution in https://github.com/microsoft/DeepSpeed-MII/pull/53

**Full Changelog**: https://github.com/microsoft/DeepSpeed-MII/commits/v0.0.2