@@ -15,7 +15,7 @@ If this is your first time, please read our contributor guidelines: https://gite | |||||
**Which issue(s) this PR fixes**: | **Which issue(s) this PR fixes**: | ||||
<!-- | |||||
<!-- | |||||
*Automatically closes linked issue when PR is merged. | *Automatically closes linked issue when PR is merged. | ||||
Usage: `Fixes #<issue number>`, or `Fixes (paste link of issue)`. | Usage: `Fixes #<issue number>`, or `Fixes (paste link of issue)`. | ||||
--> | --> | ||||
@@ -14,7 +14,7 @@ If this is your first time, please read our contributor guidelines: https://gith | |||||
**Which issue(s) this PR fixes**: | **Which issue(s) this PR fixes**: | ||||
<!-- | |||||
<!-- | |||||
*Automatically closes linked issue when PR is merged. | *Automatically closes linked issue when PR is merged. | ||||
Usage: `Fixes #<issue number>`, or `Fixes (paste link of issue)`. | Usage: `Fixes #<issue number>`, or `Fixes (paste link of issue)`. | ||||
--> | --> | ||||
@@ -6,6 +6,13 @@ the project. SIGs have open and transparent proceedings to develop goals and | |||||
implement code contributions. SIGs are also responsible for ongoing maintenance | implement code contributions. SIGs are also responsible for ongoing maintenance | ||||
of the code in their areas. | of the code in their areas. | ||||
## Proposing a new SIG | |||||
New SIGs are created when there is sufficient interest in a topic area | |||||
and someone volunteers to be the lead for the group and submits a proposal to | |||||
the steering committee. The chair facilitates the discussion and helps | |||||
synthesize proposals and decisions. | |||||
## Joining a SIG | ## Joining a SIG | ||||
If you are interested in participating, please [join the discussion](https://mailweb.mindspore.cn/postorius/lists/) | If you are interested in participating, please [join the discussion](https://mailweb.mindspore.cn/postorius/lists/) | ||||
@@ -16,16 +23,16 @@ in the mailing list. SIG artifacts can be found in the current repository. | |||||
| SIG name | Responsibilities | SIG Leads | | | SIG name | Responsibilities | SIG Leads | | ||||
| :------- | :--------------- | :-------- | | | :------- | :--------------- | :-------- | | ||||
| FrontEnd | This SIG is responsible for the development of MindSpore front-end expression. | [@kingxian](https://gitee.com/kingxian)(jianxiaoxian@huawei.com) | | |||||
| Compiler | This SIG is responsible for the development of MindSpore high level graph compilation. | [@zh_qh](https://gitee.com/zh_qh)(zhangqinghua3@huawei.com) | | |||||
| Executor | This SIG is responsible for the development of MindSpore back-end support for pipeline. | [@kisnwang](https://gitee.com/kisnwang)(wangkaisheng2@huawei.com) | | |||||
| ModelZoo | This SIG is responsible for the development of MindSpore modelzoo and additional ops. | [@yingjy](https://gitee.com/yingjy)(yingjiangyong@huawei.com) | | |||||
| Data | This SIG is responsible for the development of MindSpore data processing and data format transformation. | [@liucunwei](https://gitee.com/liucunwei)(liucunwei@huawei.com) | | |||||
| GraphEngine | This SIG is responsible for the development of MindSpore graph engine for Ascend AI processor. | [@youui](https://gitee.com/youui)(yangwei79@huawei.com) | | |||||
| Visualization | This SIG is responsible for the development of MindSpore visualization tools. | [@gaocongli_hw](https://gitee.com/gaocongli_hw)(gaocongli@huawei.com) | | |||||
| Security | This SIG is responsible for the development of MindSpore security related tools. | [@randywangze](https://gitee.com/randywangze)(wangze14@huawei.com) | | |||||
| AKG | This SIG is responsible for the development of MindSpore auto kernel generator. | [@anyrenwei](https://gitee.com/anyrenwei)(zhangrenwei1@huawei.com) | | |||||
| MSLITE | This SIG is responsible for the development of MindSpore lite. | [@zhaizhiqiang](https://gitee.com/zhaizhiqiang)(zhaizhiqiang@huawei.com) | | |||||
| MDP | This SIG is responsible for the development of MindSpore programming library for Bayesian deep learning. | [@jianfeichen](https://gitee.com/jianfeichen)(chrisjianfeichen@gmail.com) | | |||||
| Parallel | This SIG is responsible for the development of MindSpore's functionality of automatically finding the efficient parallel strategy for DNN training and inference. | [@dr-orange](https://gitee.com/dr-orange)(chengli7@ustc.edu.cn) | | |||||
| Serving | This SIG is responsible for the development of MindSpore Serving module. | [@xu-yfei](https://gitee.com/xu-yfei)(xuyongfei@huawei.com) | | |||||
| [FrontEnd](frontend/README.md) | This SIG is responsible for the development of MindSpore front-end expression. | [@kingxian](https://gitee.com/kingxian)(jianxiaoxian@huawei.com) | | |||||
| [Compiler](compiler/README.md) | This SIG is responsible for the development of MindSpore high level graph compilation. | [@zh_qh](https://gitee.com/zh_qh)(zhangqinghua3@huawei.com) | | |||||
| [Executor](executor/README.md) | This SIG is responsible for the development of MindSpore back-end support for pipeline. | [@kisnwang](https://gitee.com/kisnwang)(wangkaisheng2@huawei.com) | | |||||
| [ModelZoo](modelzoo/README.md) | This SIG is responsible for the development of MindSpore modelzoo and additional ops. | [@yingjy](https://gitee.com/yingjy)(yingjiangyong@huawei.com) | | |||||
| [Data](data/README.md) | This SIG is responsible for the development of MindSpore data processing and data format transformation. | [@liucunwei](https://gitee.com/liucunwei)(liucunwei@huawei.com) | | |||||
| [GraphEngine](graphengine/README.md) | This SIG is responsible for the development of MindSpore graph engine for Ascend AI processor. | [@youui](https://gitee.com/youui)(yangwei79@huawei.com) | | |||||
| [Visualization](visualization/README.md) | This SIG is responsible for the development of MindSpore visualization tools. | [@gaocongli_hw](https://gitee.com/gaocongli_hw)(gaocongli@huawei.com) | | |||||
| [Security](security/README.md) | This SIG is responsible for the development of MindSpore security related tools. | [@randywangze](https://gitee.com/randywangze)(wangze14@huawei.com) | | |||||
| [AKG](akg/README.md) | This SIG is responsible for the development of MindSpore auto kernel generator. | [@anyrenwei](https://gitee.com/anyrenwei)(zhangrenwei1@huawei.com) | | |||||
| [MSLITE](mslite/README.md) | This SIG is responsible for the development of MindSpore lite. | [@zhaizhiqiang](https://gitee.com/zhaizhiqiang)(zhaizhiqiang@huawei.com) | | |||||
| [MDP](mdp/README.md) | This SIG is responsible for the development of MindSpore programming library for Bayesian deep learning. | [@jianfeichen](https://gitee.com/jianfeichen)(chrisjianfeichen@gmail.com) | | |||||
| [Parallel](parallel/README.md) | This SIG is responsible for the development of MindSpore's functionality of automatically finding the efficient parallel strategy for DNN training and inference. | [@dr-orange](https://gitee.com/dr-orange)(chengli7@ustc.edu.cn) | | |||||
| [Serving](serving/README.md) | This SIG is responsible for the development of MindSpore Serving module. | [@xu-yfei](https://gitee.com/xu-yfei)(xuyongfei@huawei.com) | |
@@ -1,15 +1,19 @@ | |||||
# MindSpore Adaptive Distributed Training System Special Interest Group (SIG) | # MindSpore Adaptive Distributed Training System Special Interest Group (SIG) | ||||
The elastic systems allow users to dynamically change the number of GPUs allocated to training jobs. The target of this SIG is to develop an adaptive distributed training system that can train the neural networks in elastic clusters without affecting the convergence. This working repo contains all the airtifacts, materials, meeting notes, and proposals regarding **Elastic Training** and **Adaptive Training**. Feedbacks and contributions are welcomed. | |||||
1. **Elastic Training**: the number of GPUs could change without interrupting the traning process. | |||||
The elastic systems allow users to dynamically change the number of GPUs allocated to training jobs. The target of this SIG is to develop an adaptive distributed training system that can train the neural networks in elastic clusters without affecting the convergence. This working repo contains all the artifacts, materials, meeting notes, and proposals regarding **Elastic Training** and **Adaptive Training**. Feedbacks and contributions are welcomed. | |||||
1. **Elastic Training**: the number of GPUs could change without interrupting the training process. | |||||
2. **Adaptive Training**: the training jobs could be reconfigured and scheduled adaptively when the training resource changes, so that the convergence speed is not affected. | 2. **Adaptive Training**: the training jobs could be reconfigured and scheduled adaptively when the training resource changes, so that the convergence speed is not affected. | ||||
## SIG Leads | ## SIG Leads | ||||
Luo Mai (University of Edinburgh) | Luo Mai (University of Edinburgh) | ||||
## Logistics | ## Logistics | ||||
- SIG leads will drive the meeting. | - SIG leads will drive the meeting. | ||||
- Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/adaptivetraining | |||||
- Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/adaptivetraining | |||||
- Feedbacks and topic requests are welcomed by all. | - Feedbacks and topic requests are welcomed by all. | ||||
## Discussion | ## Discussion | ||||
- Slack channel: https://app.slack.com/client/T018BLCMSGL/learning-slack | - Slack channel: https://app.slack.com/client/T018BLCMSGL/learning-slack | ||||
- Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/adaptivetraining | - Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/adaptivetraining | ||||
## Meeting notes | ## Meeting notes |
@@ -13,7 +13,7 @@ This is the working repo for the AKG Special Interest Group (SIG). This repo con | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/akg | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/akg | |||||
* Feedbacks and topic requests are welcomed by all. | * Feedbacks and topic requests are welcomed by all. | ||||
# Discussion | # Discussion | ||||
@@ -2,7 +2,7 @@ | |||||
This is the working repo for the Data special interest group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding **dataset - data processing** and **mindrecord - data format** in MindSpore. Feedbacks and contributions are welcome. | This is the working repo for the Data special interest group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding **dataset - data processing** and **mindrecord - data format** in MindSpore. Feedbacks and contributions are welcome. | ||||
1. **Data Processing**: You can understand it as a Dataset, which is mainly responsible for reading the user's data into a Dataset, then performing related data enhancement operations (such as: resize, onehot, rotate, shuffle, batch ...), and finally provide the Dataset to the training process. | 1. **Data Processing**: You can understand it as a Dataset, which is mainly responsible for reading the user's data into a Dataset, then performing related data enhancement operations (such as: resize, onehot, rotate, shuffle, batch ...), and finally provide the Dataset to the training process. | ||||
2. **Data Format**: It can conveniently normalize the user's training data to a unified format (MindRecord). The specific operation steps are as follows: The user can easily convert the training data into MindRecrod data by defining the training data schema and calling the Python API interface. The format is then read into a Dataset through MindDataset and provided to the training process. | |||||
2. **Data Format**: It can conveniently normalize the user's training data to a unified format (MindRecord). The specific operation steps are as follows: The user can easily convert the training data into MindRecord data by defining the training data schema and calling the Python API interface. The format is then read into a Dataset through MindDataset and provided to the training process. | |||||
# SIG Leads | # SIG Leads | ||||
@@ -11,12 +11,12 @@ This is the working repo for the Data special interest group (SIG). This repo co | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/data | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/data | |||||
* Feedbacks and topic requests are welcome by all. | * Feedbacks and topic requests are welcome by all. | ||||
# Discussion | # Discussion | ||||
* Slack channel https://app.slack.com/client/TUKCY4QDR/C010RPN6QNP?cdn_fallback=2 | |||||
* Slack channel: https://app.slack.com/client/TUKCY4QDR/C010RPN6QNP?cdn_fallback=2 | |||||
* Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/data | * Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/data | ||||
# Meeting notes | # Meeting notes | ||||
@@ -1,9 +1,9 @@ | |||||
# MindSpore Executor Special Interest Group (SIG) | # MindSpore Executor Special Interest Group (SIG) | ||||
This is the working repo for the Executor Special Interest Group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding **graph optimizing** , **graph compling**, **hardware runtime**, **diagnostics**, and any other backend programs in MindSpore. Feedbacks and contributions are welcomed. | |||||
This is the working repo for the Executor Special Interest Group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding **graph optimizing** , **graph compiling**, **hardware runtime**, **diagnostics**, and any other backend programs in MindSpore. Feedbacks and contributions are welcomed. | |||||
1. **Graph Optimizing**: Similar to graph "complier", there are many Passes to fuse operators and optimize graphs. | 1. **Graph Optimizing**: Similar to graph "complier", there are many Passes to fuse operators and optimize graphs. | ||||
2. **Graph Compling**: Allocating memory and other resources, compling graphs to generate runnable models. | |||||
3. **Hardware Runtime**: Interface for hardware, who supportes ascend910, gpu and cpu now. | |||||
2. **Graph Compiling**: Allocating memory and other resources, compiling graphs to generate runnable models. | |||||
3. **Hardware Runtime**: Interface for hardware, who supports ascend910, gpu and cpu now. | |||||
4. **Diagnostics**: Provide some tools such as profiling, exporting graph ir and exporting operator data. | 4. **Diagnostics**: Provide some tools such as profiling, exporting graph ir and exporting operator data. | ||||
# SIG Leads | # SIG Leads | ||||
@@ -13,7 +13,7 @@ This is the working repo for the Executor Special Interest Group (SIG). This rep | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/executor | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/executor | |||||
* Feedbacks and topic requests are welcomed by all. | * Feedbacks and topic requests are welcomed by all. | ||||
# Discussion | # Discussion | ||||
@@ -13,7 +13,7 @@ This is the working repo for the FrontEnd Special Interest Group (SIG). This rep | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/frontend | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/frontend | |||||
* Feedbacks and topic requests are welcomed by all. | * Feedbacks and topic requests are welcomed by all. | ||||
# Discussion | # Discussion | ||||
@@ -11,7 +11,7 @@ This is the working repo for the GraphEngine Special Interest Group (SIG). This | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/graphengine | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/graphengine | |||||
* Feedbacks and topic requests are welcome by all. | * Feedbacks and topic requests are welcome by all. | ||||
# Discussion | # Discussion | ||||
@@ -21,4 +21,5 @@ This is the working repo for the GraphEngine Special Interest Group (SIG). This | |||||
# Meeting notes | # Meeting notes | ||||
* [Friday May 15, 2020](./meetings/001-20200515.md) | |||||
* [Friday July 31, 2020](./meetings/002-20200731.md) |
@@ -1,7 +1,7 @@ | |||||
# Friday July 31, 2020 at 11:00am GMT+8 | # Friday July 31, 2020 at 11:00am GMT+8 | ||||
## Agenda | ## Agenda | ||||
* GraphEngine dynamic shape feature code review | |||||
* GraphEngine dynamic shape feature code review | |||||
## Conference links | ## Conference links | ||||
* https://meet.jit.si/MindSporeGraphEngine | * https://meet.jit.si/MindSporeGraphEngine | ||||
@@ -1,16 +1,20 @@ | |||||
# MindSpore MDP Special Interest Group (SIG) | # MindSpore MDP Special Interest Group (SIG) | ||||
This is the working repo for the MDP Special Interest Group (SIG). MindSpore Deep Probabilistic Programming (MDP) is a programming library for Bayesian deep learning. The target of MDP is to intergrade the gap between deep learning and Bayesian learning. This repo contains all the artifacts, materials, meeting notes and proposals regarding **Probabilistic Programming** , **Deep Probabilistic Programming** , **Toolbox** . Feedbacks and contributions are welcomed. | This is the working repo for the MDP Special Interest Group (SIG). MindSpore Deep Probabilistic Programming (MDP) is a programming library for Bayesian deep learning. The target of MDP is to intergrade the gap between deep learning and Bayesian learning. This repo contains all the artifacts, materials, meeting notes and proposals regarding **Probabilistic Programming** , **Deep Probabilistic Programming** , **Toolbox** . Feedbacks and contributions are welcomed. | ||||
1. Probabilistic Programming: Probabilistic Programming (PP) focuses on professional Bayesian learning, incluiding statistical distributions classes used to generate stochastic tensors and probabilistic inference algorithms. | |||||
1. Probabilistic Programming: Probabilistic Programming (PP) focuses on professional Bayesian learning, including statistical distributions classes used to generate stochastic tensors and probabilistic inference algorithms. | |||||
2. Deep Probabilistic Programming: Deep Probabilistic Programming (DPP) aims to provide composable BNN modules, which contains bnn layers, bnn modules, transforms and context. | 2. Deep Probabilistic Programming: Deep Probabilistic Programming (DPP) aims to provide composable BNN modules, which contains bnn layers, bnn modules, transforms and context. | ||||
3. Toolbox: Toolbox provides a set of BNN tools for some specific applications, sunch as Uncertainty Estimation, OoD Detection and so on. | |||||
3. Toolbox: Toolbox provides a set of BNN tools for some specific applications, such as Uncertainty Estimation, OoD Detection and so on. | |||||
## SIG Leads | ## SIG Leads | ||||
Chen Jianfei (Tsinghua University) | Chen Jianfei (Tsinghua University) | ||||
## Logistics | ## Logistics | ||||
- SIG leads will drive the meeting. | - SIG leads will drive the meeting. | ||||
- Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/mdp | |||||
- Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/mdp | |||||
- Feedbacks and topic requests are welcomed by all. | - Feedbacks and topic requests are welcomed by all. | ||||
## Discussion | ## Discussion | ||||
- Slack channel: https://app.slack.com/client/T018BLCMSGL/learning-slack | - Slack channel: https://app.slack.com/client/T018BLCMSGL/learning-slack | ||||
- Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/mdp | - Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/mdp | ||||
## Meeting notes | ## Meeting notes |
@@ -0,0 +1,14 @@ | |||||
# Thursday April 2, 2020 at 21:30pm GMT+8 | |||||
## Agenda | |||||
## Conference links | |||||
## Attendees | |||||
* Tom (Huawei) | |||||
## Notes | |||||
* TODO | |||||
## Action items | |||||
* TODO |
@@ -1,8 +1,8 @@ | |||||
# MindSpore ModelZoo Special Interest Group (SIG) | # MindSpore ModelZoo Special Interest Group (SIG) | ||||
This is the working repo for the ModelZoo special interest group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding **state-of-the-art deep learning models** and **implementations** in MindSpore. Feedbacks and contributions are welcome. | This is the working repo for the ModelZoo special interest group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding **state-of-the-art deep learning models** and **implementations** in MindSpore. Feedbacks and contributions are welcome. | ||||
1. **State-of-the-Art Deep Learning Models**: It covers typical deep learning models in image classification, object detection and segmentation, and natural language processing. These models are intended to be well-maintained, tested and kept up to date with the latest Mindspore API. | |||||
2. **Implementations**: It provides a collection of example implementations for the models powered by Mindspore high-level APIs. Before implementing the model, make sure that the operations used in the model architecture and data processing pipeline are supported in Mindspore. Users can choose the related model to perform end-to-end training and do evaluation on new dataset. | |||||
1. **State-of-the-Art Deep Learning Models**: It covers typical deep learning models in image classification, object detection and segmentation, and natural language processing. These models are intended to be well-maintained, tested and kept up to date with the latest MindSpore API. | |||||
2. **Implementations**: It provides a collection of example implementations for the models powered by MindSpore high-level APIs. Before implementing the model, make sure that the operations used in the model architecture and data processing pipeline are supported in MindSpore. Users can choose the related model to perform end-to-end training and do evaluation on new dataset. | |||||
# SIG Leads | # SIG Leads | ||||
@@ -11,12 +11,12 @@ This is the working repo for the ModelZoo special interest group (SIG). This rep | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/modelzoo | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/modelzoo | |||||
* Feedbacks and topic requests are welcome by all. | * Feedbacks and topic requests are welcome by all. | ||||
# Discussion | # Discussion | ||||
* Slack channel https://app.slack.com/client/TUKCY4QDR/C010RPN6QNP?cdn_fallback=2 | |||||
* Slack channel: https://app.slack.com/client/TUKCY4QDR/C010RPN6QNP?cdn_fallback=2 | |||||
* Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/modelzoo | * Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/modelzoo | ||||
# Meeting notes | # Meeting notes | ||||
@@ -1,6 +1,7 @@ | |||||
# MindSpore Lite Special Interest Group (SIG) | # MindSpore Lite Special Interest Group (SIG) | ||||
This is the working repo for the mslite Special Interest Group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding **MS Lite Converter** , **MS Lite Runtime**. Feedbacks and contributions are welcomed. | This is the working repo for the mslite Special Interest Group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding **MS Lite Converter** , **MS Lite Runtime**. Feedbacks and contributions are welcomed. | ||||
1. **Converter**: converter is an offline tool has three parts, frontend, IR, and backend, aims to generate a compact model with applying graph optimizations and post training quantization. | 1. **Converter**: converter is an offline tool has three parts, frontend, IR, and backend, aims to generate a compact model with applying graph optimizations and post training quantization. | ||||
2. **Runtime**: runtime deploys to device and executes online, has Lite RT and Lite Micro two modes. | 2. **Runtime**: runtime deploys to device and executes online, has Lite RT and Lite Micro two modes. | ||||
@@ -11,7 +12,7 @@ This is the working repo for the mslite Special Interest Group (SIG). This repo | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/mslite | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/mslite | |||||
* Feedbacks and topic requests are welcomed by all. | * Feedbacks and topic requests are welcomed by all. | ||||
# Discussion | # Discussion | ||||
@@ -1,6 +1,7 @@ | |||||
# MindSpore Parallel Special Interest Group (SIG) | # MindSpore Parallel Special Interest Group (SIG) | ||||
This is the working repository for the Parallel Special Interest Group (SIG). This repository contains all the artifacts, materials, meeting notes and proposals regarding **Auto-parallel**, **Model-parallel**, **Pipelined model-parallel**, **Tensor partitioning**, **Cost model**. Feedback and contributions are welcome. | This is the working repository for the Parallel Special Interest Group (SIG). This repository contains all the artifacts, materials, meeting notes and proposals regarding **Auto-parallel**, **Model-parallel**, **Pipelined model-parallel**, **Tensor partitioning**, **Cost model**. Feedback and contributions are welcome. | ||||
1. **Auto-parallel**: The sizes of popular DNN models are getting larger, thus it is desired to automatically find an efficient way to parallelize the execution (training and inference) of the giant DNNs. This is the ultimate goal of this SIG. | 1. **Auto-parallel**: The sizes of popular DNN models are getting larger, thus it is desired to automatically find an efficient way to parallelize the execution (training and inference) of the giant DNNs. This is the ultimate goal of this SIG. | ||||
2. **Model-parallel**: Unlike Data-parallel in which each device holds the entire model in training, Model-parallel is to partition the model to available devices, so that each device holds a slice of the entire model. Model-parallel is a more suitable approach for training giant models. | 2. **Model-parallel**: Unlike Data-parallel in which each device holds the entire model in training, Model-parallel is to partition the model to available devices, so that each device holds a slice of the entire model. Model-parallel is a more suitable approach for training giant models. | ||||
3. **Pipelined model-parallel**: This is a paradigm to implement Model-parallel. This paradigm is to assign operators of a DNN model to different devices, so that different training batches can be pipelined. | 3. **Pipelined model-parallel**: This is a paradigm to implement Model-parallel. This paradigm is to assign operators of a DNN model to different devices, so that different training batches can be pipelined. | ||||
@@ -13,7 +14,7 @@ This is the working repository for the Parallel Special Interest Group (SIG). Th | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/parallel | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/parallel | |||||
* Feedbacks and topic requests are welcomed by all. | * Feedbacks and topic requests are welcomed by all. | ||||
# Discussion | # Discussion | ||||
@@ -21,4 +22,4 @@ This is the working repository for the Parallel Special Interest Group (SIG). Th | |||||
* Slack channel: https://app.slack.com/client/TUKCY4QDR/CUZ3FESNS?cdn_fallback=2 | * Slack channel: https://app.slack.com/client/TUKCY4QDR/CUZ3FESNS?cdn_fallback=2 | ||||
* Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/parallel | * Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/parallel | ||||
# Meeting notes | |||||
# Meeting notes |
@@ -12,7 +12,7 @@ This is the working repo for the MindArmour special interest group (SIG). This r | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/security | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/security | |||||
* Feedbacks and topic requests are welcome by all. | * Feedbacks and topic requests are welcome by all. | ||||
# Discussion | # Discussion | ||||
@@ -1,7 +1,8 @@ | |||||
# MindSpore Serving Special Interest Group (SIG) | # MindSpore Serving Special Interest Group (SIG) | ||||
This is the working repo for the Serving Special Interest Group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding. Feedbacks and contributions are welcomed. | This is the working repo for the Serving Special Interest Group (SIG). This repo contains all the artifacts, materials, meeting notes and proposals regarding. Feedbacks and contributions are welcomed. | ||||
1. Adjust and enhance the external interface: grpc + restful. | |||||
1. Adjust and enhance the external interface: gRPC + RESTful. | |||||
2. Single model concurrent, give full play to chip capability. | 2. Single model concurrent, give full play to chip capability. | ||||
3. The model implements two forms of batch. | 3. The model implements two forms of batch. | ||||
4. Support data pre-processing and post-processing sink to the service back-end. | 4. Support data pre-processing and post-processing sink to the service back-end. | ||||
@@ -14,12 +15,12 @@ This is the working repo for the Serving Special Interest Group (SIG). This repo | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/serving | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/serving | |||||
* Feedbacks and topic requests are welcomed by all. | * Feedbacks and topic requests are welcomed by all. | ||||
# Discussion | # Discussion | ||||
* Slack channel: | |||||
* Slack channel: *TBD* | |||||
* Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/serving | * Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/serving | ||||
# Meeting notes | # Meeting notes | ||||
@@ -9,12 +9,12 @@ This is the working repo for the `Visualization` special interest group (SIG). T | |||||
# Logistics | # Logistics | ||||
* SIG leads will drive the meeting. | * SIG leads will drive the meeting. | ||||
* Meeting annoucement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/visualization | |||||
* Meeting announcement will be posted on our gitee channel: https://gitee.com/mindspore/community/tree/master/sigs/visualization | |||||
* Feedbacks and topic requests are welcome by all. | * Feedbacks and topic requests are welcome by all. | ||||
# Discussion | # Discussion | ||||
* Slack channel https://app.slack.com/client/TUKCY4QDR/C0119H4M5KJ/details/members?cdn_fallback=2 | |||||
* Slack channel: https://app.slack.com/client/TUKCY4QDR/C0119H4M5KJ/details/members?cdn_fallback=2 | |||||
* Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/visualization | * Documents and artifacts: https://gitee.com/mindspore/community/tree/master/sigs/visualization | ||||
# Meeting notes | # Meeting notes | ||||
@@ -26,6 +26,6 @@ Working Group artifacts can be found in the current repository. | |||||
| WG name | Responsibilities | WG Leads | | | WG name | Responsibilities | WG Leads | | ||||
| :------ | :--------------- | :------- | | | :------ | :--------------- | :------- | | ||||
| Research | <center>—</center> | <center>—</center> | | |||||
| MM | Molecular Modeling WG aims to build a community collaboration for deep learning framework's application in molecular modeling and simulation. | [@helloyesterday](https://gitee.com/helloyesterday) [@jz_90](https://gitee.com/jz_90) | | |||||
| DOC | Discuss for improving MindSpore documentation. |[@leiyuning](https://gitee.com/jonahlei)| | |||||
| [Research](research/README.md) | <center>—</center> | <center>—</center> | | |||||
| [MM](mm/README.md) | Molecular Modeling WG aims to build a community collaboration for deep learning framework's application in molecular modeling and simulation. | [@helloyesterday](https://gitee.com/helloyesterday) [@jz_90](https://gitee.com/jz_90) | | |||||
| [Doc](doc/README.md) | Discussion for improving MindSpore documentation. |[@leiyuning](https://gitee.com/jonahlei)| |