push Deepspeed 0.6.3 rocm version

CONTRIBUTING.md

@@ -65,3 +65,31 @@ Conduct](https://opensource.microsoft.com/codeofconduct/). For more information
 [Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/) or contact
 [opencode@microsoft.com](mailto:opencode@microsoft.com) with any additional questions or
 comments.
+
+## New Feature Contribution Guidelines
+Unlike bug fix or improving existing feature (where users usually directly submit a PR and we review it), adding a new feature to DeepSpeed requires several steps: (1) proposal and discussion, (2) implementation and verification, (3) release and maintenance. This general guideline applies to all new feature contributions. Core DeepSpeed team member contributions may complete step 1 internally.
+
+### Step 1: proposal and discussion
+We ask users to first post your intended feature in an issue. This issue needs to include:
+
+* A description of the proposed feature.
+* A motivation of why it will be useful to DeepSpeed users.
+* A rough design of how you implement the feature inside DeepSpeed.
+* (Important) Results or planned experiments to demonstrate the effectiveness and correctness of the feature.
+  * If this is a general feature applicable to different tasks, we require testing it on at least one CV task (e.g., [CIFAR](https://www.deepspeed.ai/tutorials/cifar-10/)) and one NLP task (e.g., [SQuAD](https://www.deepspeed.ai/tutorials/bert-finetuning/)). If this is a feature for one kind of task only, it is fine to just test on the specific task.
+  * If the feature only affects performance and does not affect training convergence, we require testing on a fraction of training to demonstrate that the training/validation loss are consistent with baseline, and that the performance is better than baseline.
+  * If the feature does affect training convergence, we require testing the whole training to demonstrate that the feature achieves better/on-par final model quality and training performance compared to baseline.
+
+Based on the issue we shall discuss the merit of the new feature and decide whether accept or decline the proposal. Once accepted and after we confirm the design and implementation plan, we are ready for step 2.
+
+### Step 2: implementation and verification
+Contributor will go ahead and implement the feature, and the DeepSpeed team will provide guidance/helps as needed. The required deliverables include:
+
+* A PR to [microsoft/DeepSpeed](https://github.com/microsoft/DeepSpeed) including (1) the feature implementation (2) unit tests (3) documentation (4) tutorial
+* A PR to [microsoft/DeepSpeedExamples](https://github.com/microsoft/DeepSpeedExamples) or [microsoft/Megatron-DeepSpeed](https://github.com/microsoft/Megatron-DeepSpeed) including the examples of how to use the feature (this is related to the planned testing experiments in proposal)
+* In the implementation (code, documentation, tutorial), we require the feature author to record their GitHub username as a contact method for future questions/maintenance.
+
+After receiving the PRs, we will review them and merge them after necessary tests/fixes.
+
+### Step 3: release and maintenance
+After the PRs are merged, we will announce the feature on our website (with credit to the feature author). We ask the feature author to commit to the maintenance of the feature.

MANIFEST.in

 include *.txt README.md
 recursive-include requirements *.txt
-recursive-include deepspeed *.cpp *.h *.cu *.tr *.cuh *.cc
+recursive-include deepspeed *.cpp *.h *.cu *.hip *.tr *.cuh *.cc *.json
 recursive-include csrc *.cpp *.h *.cu *.tr *.cuh *.cc

MANIFEST_win.in

+include *.txt README.md
+recursive-include requirements *.txt
+
+# this is for Windows only
+recursive-include deepspeed *.tr
+recursive-exclude deepspeed/ops/csrc *.cpp *.h *.cu *.cuh *.cc
+prune csrc
+prune op_builder

README.md

@@ -2,9 +2,28 @@
 [![PyPI version](https://badge.fury.io/py/deepspeed.svg)](https://pypi.org/project/deepspeed/)
 [![Documentation Status](https://readthedocs.org/projects/deepspeed/badge/?version=latest)](https://deepspeed.readthedocs.io/en/latest/?badge=latest)
 [![License MIT](https://img.shields.io/badge/License-MIT-blue.svg)](https://github.com/Microsoft/DeepSpeed/blob/master/LICENSE)
-[![Docker Pulls](https://img.shields.io/docker/pulls/deepspeed/deepspeed)](https://hub.docker.com/r/deepspeed/deepspeed)
 
-### 03/2021: DeepSpeed is hiring! Come join us: [SDE 2](https://careers.microsoft.com/us/en/job/1013160/Software-Engineer-2), [Sr. SDE](https://careers.microsoft.com/us/en/job/1017151/Senior-Software-Engineer), [Sr. Researcher](https://careers.microsoft.com/us/en/job/1016440/Senior-Researcher)
+<div align="center">
+ <img src="docs/assets/images/DeepSpeed_light.svg#gh-light-mode-only" width="400px">
+ <img src="docs/assets/images/DeepSpeed_dark_transparent.svg#gh-dark-mode-only" width="400px">
+</div>
+
+<!--
+Remove until pypi issue is resolved: https://status.python.org/incidents/2jj696st6yn5
+[![Downloads](https://pepy.tech/badge/deepspeed/month)](https://pepy.tech/project/deepspeed)
+-->
+## Latest News
+* [2022/03/21] [Supporting efficient large model training on AMD Instinct GPUs with DeepSpeed](https://cloudblogs.microsoft.com/opensource/2022/03/21/supporting-efficient-large-model-training-on-amd-instinct-gpus-with-deepspeed/)
+* [2022/03/07] [Maximizing Communication Efficiency for Large-scale Training via 0/1 Adam](https://www.deepspeed.ai/tutorials/zero-one-adam/)
+* [2022/01/19] [DeepSpeed: Advancing MoE inference and training to power next-generation AI scale](https://www.microsoft.com/en-us/research/blog/deepspeed-advancing-moe-inference-and-training-to-power-next-generation-ai-scale/)
+    * [Mixture of Experts (MoE) for NLG tutorial](https://www.deepspeed.ai/tutorials/mixture-of-experts-nlg/).
+    * [Mixture of Experts (MoE) Inference tutorial](https://www.deepspeed.ai/tutorials/moe-inference-tutorial).
+* [2021/11/15] [Autotuning: Automatically discover the optimal DeepSpeed configuration that delivers good training speed](https://www.deepspeed.ai/news/2021/11/15/autotuning.html)
+* [2021/10/11] [Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, the World’s Largest and Most Powerful Generative Language Model](https://www.microsoft.com/en-us/research/blog/using-deepspeed-and-megatron-to-train-megatron-turing-nlg-530b-the-worlds-largest-and-most-powerful-generative-language-model/)
+  * Read more on how to [train large models with DeepSpeed](https://www.deepspeed.ai/tutorials/large-models-w-deepspeed/)
+
+### DeepSpeed is hiring, [come join us!](https://careers.microsoft.com/us/en/search-results?keywords=http:%2F%2Fdeepspeed.ai)
+---
 
 [DeepSpeed](https://www.deepspeed.ai/) is a deep learning optimization
 library that makes distributed training easy, efficient, and effective.
@@ -17,7 +36,7 @@ DeepSpeed delivers extreme-scale model training for everyone, from data scientis
 * Extreme scale: Using current generation of GPU clusters with hundreds of devices,  3D parallelism of DeepSpeed can efficiently train deep learning models with trillions of parameters.
 * Extremely memory efficient: With just a single GPU, ZeRO-Offload of DeepSpeed can train models with over 10B parameters, 10x bigger than the state of arts, democratizing multi-billion-parameter model training such that many deep learning scientists can explore bigger and better models.
 * Extremely long sequence length: Sparse attention of DeepSpeed powers an order-of-magnitude longer input sequence and obtains up to 6x faster execution comparing with dense transformers.
-* Extremely communication efficient: 3D parallelism improves communication efficiency allows users to train multi-billion-parameter models 2–7x faster on clusters with limited network bandwidth.  1-bit Adam reduces communication volume by up to 5x while achieving similar convergence efficiency to Adam, allowing for scaling to different types of GPU clusters and networks.
+* Extremely communication efficient: 3D parallelism improves communication efficiency allows users to train multi-billion-parameter models 2–7x faster on clusters with limited network bandwidth.  1-bit Adam, 0/1 Adam and 1-bit LAMB reduce communication volume by up to 26x while achieving similar convergence efficiency to Adam/LAMB, allowing for scaling to different types of GPU clusters and networks.
 
 Early adopters of DeepSpeed have already produced
 a language model (LM) with over 17B parameters called
@@ -31,22 +50,6 @@ information [here](https://innovation.microsoft.com/en-us/exploring-ai-at-scale)
 
 **_For further documentation, tutorials, and technical deep-dives please see [deepspeed.ai](https://www.deepspeed.ai/)!_**
 
-
-# News
-* [2021/04/01] [[DeepSpeed on AzureML] Transformers and CIFAR examples are now available on AzureML GitHub](https://github.com/Azure/azureml-examples/tree/main/workflows/train/deepspeed)
-* [2021/03/30] [[PyTorch Lightning Blog] Accessible Multi-Billion Parameter Model Training with PyTorch Lightning + DeepSpeed](https://medium.com/pytorch-lightning/accessible-multi-billion-parameter-model-training-with-pytorch-lightning-deepspeed-c9333ac3bb59)
-* [2021/03/16] [1-bit Adam v2: NCCL-based implementation and more](https://www.deepspeed.ai/tutorials/onebit-adam/)
-* [2021/03/08] [ZeRO-3 Offload: Scale your models to trillion parameters without code changes while leveraging both CPUs & GPUs](https://www.deepspeed.ai/news/2021/03/07/zero3-offload.html)
-* [2021/01/19] [[🤗Hugging Face Blog] Fit More and Train Faster With ZeRO via DeepSpeed and FairScale](https://huggingface.co/blog/zero-deepspeed-fairscale)
-* [2020/11/12] [Simplified install, JIT compiled ops, PyPI releases, and reduced dependencies](#installation)
-* [2020/11/10] [Efficient and robust compressed training through progressive layer dropping](https://www.deepspeed.ai/news/2020/10/28/progressive-layer-dropping-news.html)
-* [2020/09/10] [DeepSpeed v0.3: Extreme-scale model training for everyone](https://www.microsoft.com/en-us/research/blog/deepspeed-extreme-scale-model-training-for-everyone/)
-  * [Powering 10x longer sequences and 6x faster execution through DeepSpeed Sparse Attention](https://www.deepspeed.ai/news/2020/09/08/sparse-attention-news.html)
-  * [Training a trillion parameters with pipeline parallelism](https://www.deepspeed.ai/news/2020/09/08/pipeline-parallelism.html)
-  * [Up to 5x less communication and 3.4x faster training through 1-bit Adam](https://www.deepspeed.ai/news/2020/09/08/onebit-adam-news.html)
-  * [10x bigger model training on a single GPU with ZeRO-Offload](https://www.deepspeed.ai/news/2020/09/08/ZeRO-Offload.html)
-
-
 # Table of Contents
 | Section                                 | Description                                 |
 | --------------------------------------- | ------------------------------------------- |
@@ -96,6 +99,12 @@ If you would like to pre-install any of the DeepSpeed extensions/ops (instead
 of JIT compiling) or install pre-compiled ops via PyPI please see our [advanced
 installation instructions](https://www.deepspeed.ai/tutorials/advanced-install/).
 
+On Windows you can build wheel with following steps, currently only inference mode is supported.
+1. Install pytorch, such as pytorch 1.8 + cuda 11.1
+2. Install visual cpp build tools, such as VS2019 C++ x64/x86 build tools
+3. Launch cmd console with Administrator privilege for creating required symlink folders
+4. Run `python setup.py bdist_wheel` to build wheel in `dist` folder
+
 # Features
 Below we provide a brief feature list, see our detailed [feature
 overview](https://www.deepspeed.ai/features/) for descriptions and usage.
@@ -116,14 +125,14 @@ overview](https://www.deepspeed.ai/features/) for descriptions and usage.
 * [ZeRO-Offload](https://www.deepspeed.ai/tutorials/zero-offload/)
   * Leverage both CPU/GPU memory for model training
   * Support 10B model training on a single GPU
-* [Ultra-fast dense transformer kernels](https://www.deepspeed.ai/news/2020/05/18/bert-record.html)
-* [Sparse attention](https://www.deepspeed.ai/news/2020/09/08/sparse-attention.html)
+* [Ultra-fast dense transformer kernels](https://www.deepspeed.ai/2020/05/18/bert-record.html)
+* [Sparse attention](https://www.deepspeed.ai/2020/09/08/sparse-attention-news.html)
   * Memory- and compute-efficient sparse kernels
   * Support 10x longer sequences than dense
   * Flexible support to different sparse structures
-* [1-bit Adam](https://www.deepspeed.ai/news/2020/09/08/onebit-adam-blog-post.html)
+* [1-bit Adam](https://www.deepspeed.ai/2020/09/08/onebit-adam-blog-post.html), [0/1 Adam](https://www.deepspeed.ai/tutorials/zero-one-adam/) and [1-bit LAMB](https://www.deepspeed.ai/tutorials/onebit-lamb/)
   * Custom communication collective
-  * Up to 5x communication volume saving
+  * Up to 26x communication volume saving
 * [Additional Memory and Bandwidth Optimizations](https://www.deepspeed.ai/features/#additional-memory-and-bandwidth-optimizations)
   * Smart Gradient Accumulation
   * Communication/Computation Overlap
@@ -142,8 +151,12 @@ overview](https://www.deepspeed.ai/features/) for descriptions and usage.
   * Learning Rate Range Test
   * 1Cycle Learning Rate Schedule
 * [Simplified Data Loader](https://www.deepspeed.ai/features/#simplified-data-loader)
+* [Curriculum Learning](https://www.deepspeed.ai/tutorials/curriculum-learning/)
+  * A curriculum learning-based data pipeline that presents easier or simpler examples earlier during training
+  * Stable and 3.3x faster GPT-2 pre-training with 8x/4x larger batch size/learning rate while maintaining token-wise convergence speed
+  * Complementary to many other DeepSpeed features
 * [Performance Analysis and Debugging](https://www.deepspeed.ai/features/#performance-analysis-and-debugging)
-
+* [Mixture of Experts (MoE)](https://www.deepspeed.ai/tutorials/mixture-of-experts/)
 
 
 # Further Reading
@@ -161,7 +174,7 @@ All DeepSpeed documentation can be found on our website: [deepspeed.ai](https://
 | [Megatron-LM Tutorial](https://www.deepspeed.ai/tutorials/megatron/)                           |  Train GPT2 with DeepSpeed and Megatron-LM   |
 | [BERT Pre-training Tutorial](https://www.deepspeed.ai/tutorials/bert-pretraining/)             |  Pre-train BERT with DeepSpeed               |
 | [Learning Rate Range Test Tutorial](https://www.deepspeed.ai/tutorials/lrrt/)                  |  Faster training with large learning rates   |
-| [1Cycle Tutorial](https://www.deepspeed.ai/tutorials/1Cycle/)                                  |  SOTA learning schedule in DeepSpeed         |
+| [1Cycle Tutorial](https://www.deepspeed.ai/tutorials/one-cycle/)                               |  SOTA learning schedule in DeepSpeed         |
 
 
 
@@ -192,7 +205,12 @@ Conduct](https://opensource.microsoft.com/codeofconduct/). For more information
 2. Jeff Rasley, Samyam Rajbhandari, Olatunji Ruwase, and Yuxiong He. (2020) DeepSpeed: System Optimizations Enable Training Deep Learning Models with Over 100 Billion Parameters. [In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD '20, Tutorial)](https://dl.acm.org/doi/10.1145/3394486.3406703).
 3. Minjia Zhang, Yuxiong He. (2020) Accelerating Training of Transformer-Based Language Models with Progressive Layer Dropping. [arXiv:2010.13369](https://arxiv.org/abs/2010.13369) and [NeurIPS 2020](https://proceedings.neurips.cc/paper/2020/hash/a1140a3d0df1c81e24ae954d935e8926-Abstract.html).
 4. Jie Ren, Samyam Rajbhandari, Reza Yazdani Aminabadi, Olatunji Ruwase, Shuangyan Yang, Minjia Zhang, Dong Li, Yuxiong He. (2021) ZeRO-Offload: Democratizing Billion-Scale Model Training. [arXiv:2101.06840](https://arxiv.org/abs/2101.06840).
-5. Hanlin Tang, Shaoduo Gan, Ammar Ahmad Awan, Samyam Rajbhandari, Conglong Li, Xiangru Lian, Ji Liu, Ce Zhang, Yuxiong He. (2021) 1-bit Adam: Communication Efficient Large-Scale Training with Adam's Convergence Speed. [arXiv:2102.02888](https://arxiv.org/abs/2102.02888).
+5. Hanlin Tang, Shaoduo Gan, Ammar Ahmad Awan, Samyam Rajbhandari, Conglong Li, Xiangru Lian, Ji Liu, Ce Zhang, Yuxiong He. (2021) 1-bit Adam: Communication Efficient Large-Scale Training with Adam's Convergence Speed. [arXiv:2102.02888](https://arxiv.org/abs/2102.02888) and [ICML 2021](http://proceedings.mlr.press/v139/tang21a.html).
+6. Samyam Rajbhandari, Olatunji Ruwase, Jeff Rasley, Shaden Smith, Yuxiong He. (2021) ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning. [arXiv:2104.07857](https://arxiv.org/abs/2104.07857).
+7. Conglong Li, Ammar Ahmad Awan, Hanlin Tang, Samyam Rajbhandari, Yuxiong He. (2021) 1-bit LAMB: Communication Efficient Large-Scale Large-Batch Training with LAMB's Convergence Speed. [arXiv:2104.06069](https://arxiv.org/abs/2104.06069).
+8. Conglong Li, Minjia Zhang, Yuxiong He. (2021) Curriculum Learning: A Regularization Method for Efficient and Stable Billion-Scale GPT Model Pre-Training. [arXiv:2108.06084](https://arxiv.org/abs/2108.06084).
+9. Yucheng Lu, Conglong Li, Minjia Zhang, Christopher De Sa, Yuxiong He. (2022) Maximizing Communication Efficiency for Large-scale Training via 0/1 Adam. [arXiv:2202.06009](https://arxiv.org/abs/2202.06009).
+10. Samyam Rajbhandari, Conglong Li, Zhewei Yao, Minjia Zhang, Reza Yazdani Aminabadi, Ammar Ahmad Awan, Jeff Rasley, Yuxiong He. (2022) DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale [arXiv:2201.05596](https://arxiv.org/abs/2201.05596).
 
 # Videos
 1. DeepSpeed KDD 2020 Tutorial
@@ -206,3 +224,6 @@ Conduct](https://opensource.microsoft.com/codeofconduct/). For more information
     * Registration is free and all videos are available on-demand.
     * [ZeRO & Fastest BERT: Increasing the scale and speed of deep learning training in DeepSpeed](https://note.microsoft.com/MSR-Webinar-DeepSpeed-Registration-On-Demand.html).
 3. [DeepSpeed on AzureML](https://youtu.be/yBVXR8G8Bg8)
+4. Community Tutorials
+    * [DeepSpeed: All the tricks to scale to gigantic models](https://www.youtube.com/watch?v=pDGI668pNg0)
+    * [Turing-NLG, DeepSpeed and the ZeRO optimizer](https://www.youtube.com/watch?v=tC01FRB0M7w)

bin/deepspeed

-ds
\ No newline at end of file

bin/deepspeed

+#!/usr/bin/env python3
+
+from deepspeed.launcher.runner import main
+
+if __name__ == '__main__':
+    main()

bin/deepspeed.pt

-ds
\ No newline at end of file

bin/deepspeed.pt

+#!/usr/bin/env python3
+
+from deepspeed.launcher.runner import main
+
+if __name__ == '__main__':
+    main()

bin/ds

-#!/usr/bin/env python
+#!/usr/bin/env python3
 
 from deepspeed.launcher.runner import main