forked from mindspore-Ecosystem/mindspore
update mindspore version for modelzoo
This commit is contained in:
parent
0ef6b722c0
commit
86a0309d7b
|
|
@ -1,6 +1,7 @@
|
|||
# Contents
|
||||
|
||||
- [FCN 介绍](#FCN-介绍)
|
||||
- [Contents](#contents)
|
||||
- [FCN 介绍](#fcn-介绍)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
|
|
@ -8,6 +9,8 @@
|
|||
- [脚本介绍](#脚本介绍)
|
||||
- [脚本以及简单代码](#脚本以及简单代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [生成数据步骤](#生成数据步骤)
|
||||
- [训练数据](#训练数据)
|
||||
- [训练步骤](#训练步骤)
|
||||
- [训练](#训练)
|
||||
- [评估步骤](#评估步骤)
|
||||
|
|
@ -17,12 +20,21 @@
|
|||
- [推理过程](#推理过程)
|
||||
- [推理](#推理)
|
||||
- [模型介绍](#模型介绍)
|
||||
- [性能](#性能)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [FCN8s on PASCAL VOC 2012](#fcn8s-on-pascal-voc-2012)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [FCN8s on PASCAL VOC](#fcn8s-on-pascal-voc)
|
||||
- [如何使用](#如何使用)
|
||||
- [教程](#教程)
|
||||
- [Set context](#set-context)
|
||||
- [Load dataset](#load-dataset)
|
||||
- [Define model](#define-model)
|
||||
- [optimizer](#optimizer)
|
||||
- [loss scale](#loss-scale)
|
||||
- [callback for saving ckpts](#callback-for-saving-ckpts)
|
||||
- [随机事件介绍](#随机事件介绍)
|
||||
- [ModelZoo 主页](#ModelZoo-主页)
|
||||
- [ModelZoo 主页](#modelzoo-主页)
|
||||
|
||||
# [FCN 介绍](#contents)
|
||||
|
||||
|
|
@ -405,7 +417,7 @@ python export.py
|
|||
| Model Version | FCN-8s | FCN-8s |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | NV SMX2 V100-32G |
|
||||
| uploaded Date | 12/30/2020 (month/day/year) | 06/11/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.1.0-alpha | 1.2.0 |
|
||||
| MindSpore Version | 1.1.0 | 1.2.0 |
|
||||
| Dataset | PASCAL VOC 2012 and SBD | PASCAL VOC 2012 and SBD |
|
||||
| Training Parameters | epoch=500, steps=330, batch_size = 32, lr=0.015 | epoch=500, steps=330, batch_size = 8, lr=0.005 |
|
||||
| Optimizer | Momentum | Momentum |
|
||||
|
|
@ -424,7 +436,7 @@ python export.py
|
|||
| Model Version | FCN-8s | FCN-8s
|
||||
| Resource | Ascend 910; OS Euler2.8 | NV SMX2 V100-32G
|
||||
| Uploaded Date | 10/29/2020 (month/day/year) | 06/11/2021 (month/day/year)
|
||||
| MindSpore Version | 1.1.0-alpha | 1.2.0
|
||||
| MindSpore Version | 1.1.0 | 1.2.0
|
||||
| Dataset | PASCAL VOC 2012 | PASCAL VOC 2012
|
||||
| batch_size | 16 | 16
|
||||
| outputs | probability | probability
|
||||
|
|
|
|||
|
|
@ -1,25 +1,27 @@
|
|||
# Contents
|
||||
|
||||
- [AlexNet Description](#alexnet-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Contents](#contents)
|
||||
- [AlexNet Description](#alexnet-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [Result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## [AlexNet Description](#contents)
|
||||
|
||||
|
|
@ -406,8 +408,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| Parameters | Ascend | GPU |
|
||||
| -------------------------- | ------------------------------------------------------------| -------------------------------------------------|
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | NV SMX2 V100-32G |
|
||||
| uploaded Date | 06/09/2020 (month/day/year) | 17/09/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 | 0.7.0-beta |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) | 17/09/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.2.1 | 1.2.1 |
|
||||
| Dataset | CIFAR-10 | CIFAR-10 |
|
||||
| Training Parameters | epoch=30, steps=1562, batch_size = 32, lr=0.002 | epoch=30, steps=1562, batch_size = 32, lr=0.002 |
|
||||
| Optimizer | Momentum | Momentum |
|
||||
|
|
|
|||
|
|
@ -3,27 +3,27 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [AlexNet描述](#alexnet描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练)
|
||||
- [评估过程](#评估过程)
|
||||
- [评估](#评估)
|
||||
- [AlexNet描述](#alexnet描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练)
|
||||
- [评估过程](#评估过程)
|
||||
- [评估](#评估)
|
||||
- [推理过程](#推理过程)
|
||||
- [导出MindIR](#导出MindIR)
|
||||
- [在Ascend310执行推理](#在Ascend310执行推理)
|
||||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -332,16 +332,16 @@ bash run_infer_310.sh [MINDIR_PATH] [DATASET_NAME] [DATASET_PATH] [NEED_PREPROCE
|
|||
| 参数 | Ascend | GPU |
|
||||
| -------------------------- | ------------------------------------------------------------| -------------------------------------------------|
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 | NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-09-06 | 2020-09-17 |
|
||||
| MindSpore版本 | 0.5.0-beta | 0.7.0-beta |
|
||||
| 上传日期 | 2021-07-05 | 2020-09-17 |
|
||||
| MindSpore版本 | 1.2.1 | 1.2.1 |
|
||||
| 数据集 | CIFAR-10 | CIFAR-10 |
|
||||
| 训练参数 | epoch=30, step=1562, batch_size=32, lr=0.002 | epoch=30, step=1562, batch_size=32, lr=0.002 |
|
||||
| 优化器 | 动量 | 动量 |
|
||||
| 损失函数 | Softmax交叉熵 | Softmax交叉熵 |
|
||||
| 输出 | 概率 | 概率 | 概率 |
|
||||
| 损失 | 0.0016 | 0.01 |
|
||||
| 速度 | 21毫秒/步 | 16.8毫秒/步 |
|
||||
| 总时间 | 17分钟 | 14分钟|
|
||||
| 速度 | 7毫秒/步 | 16.8毫秒/步 |
|
||||
| 总时间 | 6分钟 | 14分钟|
|
||||
| 微调检查点 | 445M （.ckpt文件） | 445M （.ckpt文件） |
|
||||
| 脚本 | [AlexNet脚本](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/cv/alexnet) | [AlexNet脚本](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/cv/alexnet) |
|
||||
|
||||
|
|
|
|||
|
|
@ -1,6 +1,8 @@
|
|||
# Contents
|
||||
|
||||
- [DeepLabV3 Description](#DeepLabV3-description)
|
||||
- [Contents](#contents)
|
||||
- [DeepLabV3 Description](#deeplabv3-description)
|
||||
- [Description](#description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
|
|
@ -11,12 +13,26 @@
|
|||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Usage](#usage)
|
||||
- [Running on Ascend](#running-on-ascend)
|
||||
- [Running on CPU](#running-on-cpu)
|
||||
- [Result](#result)
|
||||
- [Running on Ascend](#running-on-ascend-1)
|
||||
- [Running on CPU](#running-on-cpu-1)
|
||||
- [Transfer Training](#transfer-training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Usage](#usage-1)
|
||||
- [Running on Ascend](#running-on-ascend-2)
|
||||
- [Result](#result-1)
|
||||
- [Training accuracy](#training-accuracy)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Usage](#usage-2)
|
||||
- [result](#result-2)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -803,8 +819,8 @@ Inference result is saved in current path, you can find result in acc.log file.
|
|||
| -------------------------- | -------------------------------------- |
|
||||
| Model Version | DeepLabV3
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 09/04/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.7.0-alpha |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | PASCAL VOC2012 + SBD |
|
||||
| Training Parameters | epoch = 300, batch_size = 32 (s16_r1) <br> epoch = 800, batch_size = 16 (s8_r1) <br> epoch = 300, batch_size = 16 (s8_r2) |
|
||||
| Optimizer | Momentum |
|
||||
|
|
@ -824,8 +840,8 @@ Inference result is saved in current path, you can find result in acc.log file.
|
|||
| ------------------- | --------------------------- |
|
||||
| Model Version | DeepLabV3 V1 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 09/04/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.7.0-alpha |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | VOC datasets |
|
||||
| batch_size | 32 (s16); 16 (s8) |
|
||||
| outputs | probability |
|
||||
|
|
|
|||
|
|
@ -17,10 +17,14 @@
|
|||
- [训练过程](#训练过程)
|
||||
- [用法](#用法)
|
||||
- [Ascend处理器环境运行](#ascend处理器环境运行)
|
||||
- [CPU环境运行](#cpu环境运行)
|
||||
- [迁移训练](#迁移训练)
|
||||
- [结果](#结果)
|
||||
- [Ascend处理器环境运行](#ascend处理器环境运行-1)
|
||||
- [CPU环境运行](#cpu环境运行-1)
|
||||
- [评估过程](#评估过程)
|
||||
- [用法](#用法-1)
|
||||
- [Ascend处理器环境运行](#ascend处理器环境运行-1)
|
||||
- [Ascend处理器环境运行](#ascend处理器环境运行-2)
|
||||
- [结果](#结果-1)
|
||||
- [训练准确率](#训练准确率)
|
||||
- [导出mindir模型](#导出mindir模型)
|
||||
|
|
@ -29,7 +33,8 @@
|
|||
- [结果](#结果-2)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [推理性能](#推理性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
|
|
@ -815,8 +820,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATA_PATH] [DATA_ROOT] [DATA_LIST] [DEVICE_
|
|||
| -------------------------- | -------------------------------------- |
|
||||
| 模型版本 | DeepLabV3
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-09-04 |
|
||||
| MindSpore版本 | 0.7.0-alpha |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | PASCAL VOC2012 + SBD |
|
||||
| 训练参数 | epoch = 300, batch_size = 32 (s16_r1) epoch = 800, batch_size = 16 (s8_r1) epoch = 300, batch_size = 16 (s8_r2) |
|
||||
| 优化器 | Momentum |
|
||||
|
|
@ -833,8 +838,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATA_PATH] [DATA_ROOT] [DATA_LIST] [DEVICE_
|
|||
| ------------------- | --------------------------- |
|
||||
| 模型版本 | DeepLabV3 V1 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-09-04 |
|
||||
| MindSpore 版本 | 0.7.0-alpha |
|
||||
| 上传日期 | 2020-07-05 |
|
||||
| MindSpore 版本 | 1.3.0 |
|
||||
| 数据集 | VOC 数据集 |
|
||||
| batch_size | 32 (s16); 16 (s8) |
|
||||
| 输出 | 概率 |
|
||||
|
|
|
|||
|
|
@ -2,6 +2,7 @@
|
|||
|
||||
[查看中文](./README_CN.md)
|
||||
|
||||
- [Contents](#contents)
|
||||
- [GoogleNet Description](#googlenet-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
|
|
@ -14,19 +15,23 @@
|
|||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Distributed Training](#distributed-training)
|
||||
- [Distributed Training](#distributed-training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Export Process](#Export-process)
|
||||
- [Export](#Export)
|
||||
- [Inference Process](#Inference-process)
|
||||
- [Inference](#Inference)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#evaluation-performance)
|
||||
- [How to use](#how-to-use)
|
||||
- [Export Process](#export-process)
|
||||
- [Export](#export)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Inference](#inference)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [GoogleNet on CIFAR-10](#googlenet-on-cifar-10)
|
||||
- [GoogleNet on 1200k images](#googlenet-on-1200k-images)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [GoogleNet on CIFAR-10](#googlenet-on-cifar-10-1)
|
||||
- [GoogleNet on 1200k images](#googlenet-on-1200k-images-1)
|
||||
- [How to use](#how-to-use)
|
||||
- [Inference](#inference-1)
|
||||
- [Continue Training on the Pretrained Model](#continue-training-on-the-pretrained-model)
|
||||
- [Transfer Learning](#transfer-learning)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
|
|
@ -507,8 +512,8 @@ Current batch_ Size can only be set to 1.
|
|||
| -------------------------- | ----------------------------------------------------------- | ---------------------- |
|
||||
| Model Version | Inception V1 | Inception V1 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | NV SMX2 V100-32G |
|
||||
| uploaded Date | 10/28/2020 (month/day/year) | 10/28/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | CIFAR-10 | CIFAR-10 |
|
||||
| Training Parameters | epoch=125, steps=390, batch_size = 128, lr=0.1 | epoch=125, steps=390, batch_size=128, lr=0.1 |
|
||||
| Optimizer | Momentum | Momentum |
|
||||
|
|
@ -528,8 +533,8 @@ Current batch_ Size can only be set to 1.
|
|||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| Model Version | Inception V1 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 56cores; Memory 314G; OS Euler2.8 |
|
||||
| uploaded Date | 10/28/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | 1200k images |
|
||||
| Training Parameters | epoch=300, steps=5000, batch_size=256, lr=0.1 |
|
||||
| Optimizer | Momentum |
|
||||
|
|
@ -550,8 +555,8 @@ Current batch_ Size can only be set to 1.
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| Model Version | Inception V1 | Inception V1 |
|
||||
| Resource | Ascend 910; OS Euler2.8 | GPU |
|
||||
| Uploaded Date | 10/28/2020 (month/day/year) | 10/28/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 | 1.0.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | CIFAR-10, 10,000 images | CIFAR-10, 10,000 images |
|
||||
| batch_size | 128 | 128 |
|
||||
| outputs | probability | probability |
|
||||
|
|
@ -564,8 +569,8 @@ Current batch_ Size can only be set to 1.
|
|||
| ------------------- | --------------------------- |
|
||||
| Model Version | Inception V1 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 10/28/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | 1200k images |
|
||||
| batch_size | 256 |
|
||||
| outputs | probability |
|
||||
|
|
|
|||
|
|
@ -33,7 +33,7 @@
|
|||
- [CIFAR-10上的GoogleNet](#cifar-10上的googlenet-1)
|
||||
- [120万张图像上的GoogleNet](#120万张图像上的googlenet-1)
|
||||
- [使用流程](#使用流程)
|
||||
- [推理](#推理)
|
||||
- [推理](#推理-1)
|
||||
- [继续训练预训练模型](#继续训练预训练模型)
|
||||
- [迁移学习](#迁移学习)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
|
|
@ -481,8 +481,8 @@ python export.py --config_path [CONFIG_PATH]
|
|||
| -------------------------- | ----------------------------------------------------------- | ---------------------- |
|
||||
| 模型版本 | Inception V1 | Inception V1 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 | NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-08-31 | 2020-08-20 |
|
||||
| MindSpore版本 | 0.7.0-alpha | 0.6.0-alpha |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | CIFAR-10 | CIFAR-10 |
|
||||
| 训练参数 | epoch=125, steps=390, batch_size = 128, lr=0.1 | epoch=125, steps=390, batch_size=128, lr=0.1 |
|
||||
| 优化器 | Momentum | Momentum |
|
||||
|
|
@ -502,8 +502,8 @@ python export.py --config_path [CONFIG_PATH]
|
|||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| 模型版本 | Inception V1 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，56核；内存 314G；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-09-20 |
|
||||
| MindSpore版本 | 0.7.0-alpha |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | 120万张图像 |
|
||||
| 训练参数 | epoch=300, steps=5000, batch_size=256, lr=0.1 |
|
||||
| 优化器 | Momentum |
|
||||
|
|
@ -524,8 +524,8 @@ python export.py --config_path [CONFIG_PATH]
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| 模型版本 | Inception V1 | Inception V1 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | GPU |
|
||||
| 上传日期 | 2020-08-31 | 2020-08-20 |
|
||||
| MindSpore 版本 | 0.7.0-alpha | 0.6.0-alpha |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore 版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | CIFAR-10, 1万张图像 | CIFAR-10, 1万张图像 |
|
||||
| batch_size | 128 | 128 |
|
||||
| 输出 | 概率 | 概率 |
|
||||
|
|
@ -538,8 +538,8 @@ python export.py --config_path [CONFIG_PATH]
|
|||
| ------------------- | --------------------------- |
|
||||
| 模型版本 | Inception V1 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-09-20 |
|
||||
| MindSpore版本 | 0.7.0-alpha |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | 12万张图像 |
|
||||
| batch_size | 256 |
|
||||
| 输出 | 概率 |
|
||||
|
|
|
|||
|
|
@ -1,18 +1,31 @@
|
|||
# Contents
|
||||
|
||||
- [InceptionV3 Description](#InceptionV3-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Contents](#contents)
|
||||
- [InceptionV3 Description](#inceptionv3-description)
|
||||
- [Model architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Mixed Precision(Ascend)](#mixed-precisionascend)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Training Process](#training-process)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Script description](#script-description)
|
||||
- [Script and sample code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training process](#training-process)
|
||||
- [Usage](#usage)
|
||||
- [Launch](#launch)
|
||||
- [Result](#result)
|
||||
- [Ascend](#ascend)
|
||||
- [CPU](#cpu)
|
||||
- [Eval process](#eval-process)
|
||||
- [Usage](#usage-1)
|
||||
- [Launch](#launch-1)
|
||||
- [Result](#result-1)
|
||||
- [Model Export](#model-export)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Usage](#usage-2)
|
||||
- [result](#result-2)
|
||||
- [Model description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
|
|
@ -403,8 +416,8 @@ accuracy:78.742
|
|||
| -------------------------- | ---------------------------------------------- |
|
||||
| Model Version | InceptionV3 |
|
||||
| Resource | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 08/21/2020 |
|
||||
| MindSpore Version | 0.6.0-beta |
|
||||
| uploaded Date | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | 1200k images |
|
||||
| Batch_size | 128 |
|
||||
| Training Parameters | src/model_utils/default_config.yaml |
|
||||
|
|
@ -412,11 +425,11 @@ accuracy:78.742
|
|||
| Loss Function | SoftmaxCrossEntropy |
|
||||
| Outputs | probability |
|
||||
| Loss | 1.98 |
|
||||
| Total time (8p) | 11h |
|
||||
| Total time (8p) | 10h |
|
||||
| Params (M) | 103M |
|
||||
| Checkpoint for Fine tuning | 313M |
|
||||
| Model for inference | 92M (.onnx file) |
|
||||
| Speed | 1pc:1050 img/s;8pc:8000 img/s |
|
||||
| Speed | 1pc:1200 img/s;8pc:9500 img/s |
|
||||
| Scripts | [inceptionv3 script](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/cv/inceptionv3) |
|
||||
|
||||
### Inference Performance
|
||||
|
|
@ -425,8 +438,8 @@ accuracy:78.742
|
|||
| ------------------- | --------------------------- |
|
||||
| Model Version | InceptionV3 |
|
||||
| Resource | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 |
|
||||
| Uploaded Date | 08/22/2020 |
|
||||
| MindSpore Version | 0.6.0-beta |
|
||||
| Uploaded Date | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | 50k images |
|
||||
| Batch_size | 128 |
|
||||
| Outputs | probability |
|
||||
|
|
|
|||
|
|
@ -16,10 +16,16 @@
|
|||
- [用法](#用法)
|
||||
- [启动](#启动)
|
||||
- [结果](#结果)
|
||||
- [Ascend](#ascend)
|
||||
- [CPU](#cpu)
|
||||
- [评估过程](#评估过程)
|
||||
- [用法](#用法-1)
|
||||
- [启动](#启动-1)
|
||||
- [结果](#结果-1)
|
||||
- [模型导出](#模型导出)
|
||||
- [推理过程](#推理过程)
|
||||
- [使用方法](#使用方法)
|
||||
- [结果](#结果-2)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
|
|
@ -408,8 +414,8 @@ accuracy:78.742
|
|||
| -------------------------- | ------------------------------------------------------- |
|
||||
| 模型版本 | InceptionV3 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8|
|
||||
| 上传日期 | 2020-08-21 |
|
||||
| MindSpore版本 | 0.6.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | 120万张图像 |
|
||||
| Batch_size | 128 |
|
||||
| 训练参数 | src/model_utils/default_config.yaml |
|
||||
|
|
@ -417,10 +423,10 @@ accuracy:78.742
|
|||
| 损失函数 | Softmax交叉熵 |
|
||||
| 输出 | 概率 |
|
||||
| 损失 | 1.98 |
|
||||
| 总时长（8卡） | 11小时 |
|
||||
| 总时长（8卡） | 10小时 |
|
||||
| 参数(M) | 103M |
|
||||
| 微调检查点 | 313M |
|
||||
| 训练速度 | 单卡：1050img/s;8卡：8000 img/s |
|
||||
| 训练速度 | 单卡：1200img/s;8卡：9500 img/s |
|
||||
| 脚本 | [inceptionv3脚本](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/cv/inceptionv3) |
|
||||
|
||||
#### 推理性能
|
||||
|
|
@ -429,8 +435,8 @@ accuracy:78.742
|
|||
| ------------------- | --------------------------- |
|
||||
| 模型版本 | InceptionV3 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8|
|
||||
| 上传日期 | 2020-08-22 |
|
||||
| MindSpore 版本 | 0.6.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore 版本 | 1.3.0 |
|
||||
| 数据集 | 5万张图像 |
|
||||
| Batch_size | 128 |
|
||||
| 输出 | 概率 |
|
||||
|
|
|
|||
|
|
@ -1,26 +1,28 @@
|
|||
# Contents
|
||||
|
||||
- [LeNet Description](#lenet-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Contents](#contents)
|
||||
- [LeNet Description](#lenet-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Training](#training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## [LeNet Description](#contents)
|
||||
|
||||
|
|
@ -303,8 +305,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| Parameters | LeNet |
|
||||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 09/16/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | MNIST |
|
||||
| Training Parameters | epoch=10, steps=1875, batch_size = 32, lr=0.01 |
|
||||
| Optimizer | Momentum |
|
||||
|
|
|
|||
|
|
@ -3,13 +3,13 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [LeNet描述](#lenet描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [LeNet描述](#lenet描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练)
|
||||
|
|
@ -19,12 +19,12 @@
|
|||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [推理性能](#推理性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -303,16 +303,16 @@ bash run_infer_310.sh [MINDIR_PATH] [DATA_PATH] [DVPP] [DEVICE_ID]
|
|||
| 参数 | LeNet |
|
||||
| -------------------- | ------------------------------------------------------- |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8|
|
||||
| 上传日期 | 2020-06-09 |
|
||||
| MindSpore版本 | 0.5.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | MNIST |
|
||||
| 训练参数 | epoch=10, steps=1875, batch_size = 32, lr=0.01 |
|
||||
| 优化器 | Momentum |
|
||||
| 损失函数 | Softmax交叉熵 |
|
||||
| 输出 | 概率 |
|
||||
| 损失 | 0.002 |
|
||||
| 速度 | 1.70毫秒/步 |
|
||||
| 总时长 | 43.1秒 |
|
||||
| 速度 | 1.0毫秒/步 |
|
||||
| 总时长 | 32.1秒 |
|
||||
| 微调检查点 | 482k (.ckpt文件) |
|
||||
| 脚本 | [LeNet脚本](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/cv/lenet) |
|
||||
|
||||
|
|
|
|||
|
|
@ -1,21 +1,25 @@
|
|||
# Contents
|
||||
|
||||
- [LeNet Description](#lenet-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Contents](#contents)
|
||||
- [LeNet Description](#lenet-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [Model Export](#model-export)
|
||||
- [Ascend 310 inference](#ascend-310-inference)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## [LeNet Description](#contents)
|
||||
|
||||
|
|
@ -193,9 +197,9 @@ You can view the results through the file "acc.log". The accuracy of the test da
|
|||
|
||||
| Parameters | LeNet |
|
||||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 06/09/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.5.0-beta |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | MNIST |
|
||||
| Training Parameters | epoch=10, steps=937, batch_size = 64, lr=0.01 |
|
||||
| Optimizer | Momentum |
|
||||
|
|
|
|||
|
|
@ -3,23 +3,25 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [LeNet描述](#lenet描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练)
|
||||
- [评估过程](#评估过程)
|
||||
- [评估](#评估)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [LeNet描述](#lenet描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练)
|
||||
- [评估过程](#评估过程)
|
||||
- [评估](#评估)
|
||||
- [模型导出](#模型导出)
|
||||
- [Ascend 310推理](#ascend-310推理)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -198,8 +200,8 @@ bash run_infer_310.sh [AIR_PATH] [DATA_PATH] [LABEL_PATH] [DEVICE_ID]
|
|||
| 参数 | LeNet |
|
||||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-09 |
|
||||
| MindSpore版本 | 0.5.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | MNIST |
|
||||
| 训练参数 | epoch=10, steps=937, batch_size = 64, lr=0.01 |
|
||||
| 优化器 | Momentum |
|
||||
|
|
|
|||
|
|
@ -1,23 +1,28 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [MaskRCNN Description](#maskrcnn-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Run in docker](#Run-in-docker)
|
||||
- [Run in docker](#run-in-docker)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Script Parameters](#training-script-parameters)
|
||||
- [Parameters Configuration](#parameters-configuration)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Distributed Training](#distributed-training)
|
||||
- [Training Result](#training-result)
|
||||
- [Training](#training)
|
||||
- [Distributed Training](#distributed-training)
|
||||
- [Training Result](#training-result)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Evaluation Result](#evaluation-result)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Evaluation result](#evaluation-result)
|
||||
- [Model Export](#model-export)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Usage](#usage)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
|
|
@ -706,8 +711,8 @@ Accumulating evaluation results...
|
|||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| Model Version | V1 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 08/01/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | COCO2017 |
|
||||
| Training Parameters | epoch=12, batch_size = 2 |
|
||||
| Optimizer | SGD |
|
||||
|
|
@ -727,8 +732,8 @@ Accumulating evaluation results...
|
|||
| ------------------- | --------------------------- |
|
||||
| Model Version | V1 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 08/01/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | COCO2017 |
|
||||
| batch_size | 2 |
|
||||
| outputs | mAP |
|
||||
|
|
|
|||
|
|
@ -8,6 +8,7 @@
|
|||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [在docker上运行](#在docker上运行)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本和样例代码](#脚本和样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
|
|
@ -20,12 +21,16 @@
|
|||
- [评估过程](#评估过程)
|
||||
- [评估](#评估)
|
||||
- [评估结果](#评估结果)
|
||||
- [模型导出](#模型导出)
|
||||
- [推理过程](#推理过程)
|
||||
- [使用方法](#使用方法)
|
||||
- [结果](#结果)
|
||||
- [模型说明](#模型说明)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo首页](#modelzoo首页)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -701,8 +706,8 @@ Accumulating evaluation results...
|
|||
| ------------------- | --------------------------------------------------------- |
|
||||
| 模型版本 | V1 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-08-01 |
|
||||
| MindSpore版本 | 0.6.0-alpha |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | COCO2017 |
|
||||
| 训练参数 | epoch=12，batch_size=2 |
|
||||
| 优化器 | SGD |
|
||||
|
|
@ -718,8 +723,8 @@ Accumulating evaluation results...
|
|||
| --------------------- | ----------------------------- |
|
||||
| 模型版本 | V1 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-08-01 |
|
||||
| MindSpore版本 | 0.6.0-alpha |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | COCO2017 |
|
||||
| 批次大小 | 2 |
|
||||
| 输出 | mAP |
|
||||
|
|
|
|||
|
|
@ -1,23 +1,30 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [MobileNetV2 Description](#mobilenetv2-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Model architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision(ascend))
|
||||
- [Mixed Precision(Ascend)](#mixed-precisionascend)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Training Process](#training-process)
|
||||
- [Evaluation Process](#eval-process)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Script description](#script-description)
|
||||
- [Script and sample code](#script-and-sample-code)
|
||||
- [Training process](#training-process)
|
||||
- [Usage](#usage)
|
||||
- [Launch](#launch)
|
||||
- [Result](#result)
|
||||
- [Evaluation process](#evaluation-process)
|
||||
- [Usage](#usage-1)
|
||||
- [Launch](#launch-1)
|
||||
- [Result](#result-1)
|
||||
- [Training with dataset on NFS](#training-with-dataset-on-nfs)
|
||||
- [Inference process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result-2)
|
||||
- [Model description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
|
@ -398,8 +405,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| -------------------------- | ---------------------------------------------------------- | ------------------------- |
|
||||
| Model Version | V1 | V1 |
|
||||
| Resource | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 | NV SMX2 V100-32G |
|
||||
| uploaded Date | 05/06/2020 | 05/06/2020 |
|
||||
| MindSpore Version | 0.3.0 | 0.3.0 |
|
||||
| uploaded Date | 07/05/2021 | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet | ImageNet |
|
||||
| Training Parameters | src/config.py | src/config.py |
|
||||
| Optimizer | Momentum | Momentum |
|
||||
|
|
|
|||
|
|
@ -17,10 +17,11 @@
|
|||
- [用法](#用法-1)
|
||||
- [启动](#启动-1)
|
||||
- [结果](#结果-1)
|
||||
- [NFS数据集的训练过程](#nfs数据集的训练过程)
|
||||
- [推理过程](#推理过程)
|
||||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [结果](#结果-2)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
|
|
@ -405,8 +406,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATA_PATH] [LABEL_PATH] [DVPP] [DEVICE_ID]
|
|||
| -------------------------- | ---------------------------------------------------------- | ------------------------- |
|
||||
| 模型版本 | V1 | V1 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 | NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-05-06 | 2020-05-06 |
|
||||
| MindSpore版本 | 0.3.0 | 0.3.0 |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | ImageNet | ImageNet |
|
||||
| 训练参数 | src/config.py | src/config.py |
|
||||
| 优化器 | Momentum | Momentum |
|
||||
|
|
|
|||
|
|
@ -8,6 +8,7 @@
|
|||
- [数据集](#数据集)
|
||||
- [特性](#特性)
|
||||
- [混合精度](#混合精度)
|
||||
- [量化步长可学习的量化感知训练](#量化步长可学习的量化感知训练)
|
||||
- [环境要求](#环境要求)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本和样例代码](#脚本和样例代码)
|
||||
|
|
@ -20,6 +21,8 @@
|
|||
- [用法](#用法-1)
|
||||
- [启动](#启动-1)
|
||||
- [结果](#结果-1)
|
||||
- [模型导出](#模型导出)
|
||||
- [Ascend 310 推理](#ascend-310-推理)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
|
|
@ -279,8 +282,8 @@ bash run_infer_310.sh [AIR_PATH] [DATA_PATH] [LABEL_PATH] [DEVICE_ID]
|
|||
| 量化方案 | 传统量化感知训练（默认） |量化步长可学习的量化感知训练 |
|
||||
| 量化策略 | W:8bit, A:8bit | W:4bit (首尾层为 8bit), A:8bit|
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-06 |2021-04-30 |
|
||||
| MindSpore版本 | 0.3.0 |1.3.0 |
|
||||
| 上传日期 | 2021-07-05 |2021-04-30 |
|
||||
| MindSpore版本 | 1.3.0 |1.3.0 |
|
||||
| 数据集 | ImageNet |ImageNet |
|
||||
| 训练参数 | src/config.py |src/config.py |
|
||||
| 优化器 | Momentum |Momentum |
|
||||
|
|
@ -301,8 +304,8 @@ bash run_infer_310.sh [AIR_PATH] [DATA_PATH] [LABEL_PATH] [DEVICE_ID]
|
|||
| 量化方案 | 传统量化感知训练（默认） |量化步长可学习的量化感知训练 |
|
||||
| 量化策略 | W:8bit, A:8bit | W:4bit (首尾层为 8bit), A:8bit|
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-06 |2021-04-30 |
|
||||
| MindSpore版本 | 0.3.0 | 1.3.0 |
|
||||
| 上传日期 | 2021-07-05 |2021-04-30 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | ImageNet, 1.2W | ImageNet, 1.2W |
|
||||
| 批次大小 | 130（8P） | |
|
||||
| 输出 | 概率 | 概率 |
|
||||
|
|
|
|||
|
|
@ -1,20 +1,30 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [MobileNetV2 Description](#mobilenetv2-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Model architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Learned Step Size Quantization](#learned-step-size-quantization)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script description](#script-description)
|
||||
- [Script and sample code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Model Description](#model-description)
|
||||
- [Training process](#training-process)
|
||||
- [Usage](#usage)
|
||||
- [Launch](#launch)
|
||||
- [Result](#result)
|
||||
- [Evaluation process](#evaluation-process)
|
||||
- [Usage](#usage-1)
|
||||
- [Launch](#launch-1)
|
||||
- [Result](#result-1)
|
||||
- [Model Export](#model-export)
|
||||
- [Ascend 310 inference](#ascend-310-inference)
|
||||
- [Model description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -269,8 +279,8 @@ You can view the results through the file "acc.log". The accuracy of the test da
|
|||
| Optimize Option | QAT | LEARNED_SCALE |
|
||||
| Quantization Strategy | W:8bit, A:8bit | W:4bit (The first and last layers are 8bit), A:8bit|
|
||||
| Resource | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 06/06/2020 | 04/30/2021 |
|
||||
| MindSpore Version | 0.3.0 | 1.3.0 |
|
||||
| uploaded Date | 07/05/2021 | 04/30/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet | ImageNet |
|
||||
| Training Parameters | src/config.py | src/config.py |
|
||||
| Optimizer | Momentum | Momentum |
|
||||
|
|
@ -291,8 +301,8 @@ You can view the results through the file "acc.log". The accuracy of the test da
|
|||
| Optimize Option | QAT | LEARNED_SCALE |
|
||||
| Quantization Strategy | W:8bit, A:8bit | W:4bit (The first and last layers are 8bit), A:8bit|
|
||||
| Resource | Ascend 910; OS Euler2.8 | Ascend 910; OS Euler2.8 |
|
||||
| uploaded Date | 06/06/2020 | 04/30/2021 |
|
||||
| MindSpore Version | 0.3.0 | 1.3.0 |
|
||||
| uploaded Date | 07/05/2021 | 04/30/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet, 1.2W | ImageNet, 1.2W |
|
||||
| batch_size | 130(8P) | |
|
||||
| outputs | probability | probability |
|
||||
|
|
|
|||
|
|
@ -17,7 +17,7 @@
|
|||
- [用法](#用法-1)
|
||||
- [启动](#启动-1)
|
||||
- [结果](#结果-1)
|
||||
- [导出MINDIR](#导出MINDIR)
|
||||
- [导出MINDIR](#导出mindir)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
|
|
@ -161,8 +161,8 @@ python export.py --device_target [PLATFORM] --checkpoint_path [CKPT_PATH]
|
|||
| -------------------------- | ------------------------- |
|
||||
| 模型版本 | 大版本 |
|
||||
| 资源 | NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-05-06 |
|
||||
| MindSpore版本 | 0.3.0 |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | ImageNet |
|
||||
| 训练参数 | src/config.py |
|
||||
| 优化器 | Momentum |
|
||||
|
|
|
|||
|
|
@ -1,19 +1,24 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [MobileNetV3 Description](#mobilenetv3-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Model architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Training Process](#training-process)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Script description](#script-description)
|
||||
- [Script and sample code](#script-and-sample-code)
|
||||
- [Training process](#training-process)
|
||||
- [Usage](#usage)
|
||||
- [Launch](#launch)
|
||||
- [Result](#result)
|
||||
- [Eval process](#eval-process)
|
||||
- [Usage](#usage-1)
|
||||
- [Launch](#launch-1)
|
||||
- [Result](#result-1)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#evaluation-performance)
|
||||
- [Inference Performance](#evaluation-performance)
|
||||
- [Model description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -152,8 +157,8 @@ python export.py --device_target [PLATFORM] --checkpoint_path [CKPT_PATH]
|
|||
| -------------------------- | ------------------------- |
|
||||
| Model Version | large |
|
||||
| Resource | NV SMX2 V100-32G |
|
||||
| uploaded Date | 05/06/2020 |
|
||||
| MindSpore Version | 0.3.0 |
|
||||
| uploaded Date | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | ImageNet |
|
||||
| Training Parameters | src/config.py |
|
||||
| Optimizer | Momentum |
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
# Contents
|
||||
|
||||
- [Openpose Description](#googlenet-description)
|
||||
- [Contents](#contents)
|
||||
- [Openpose Description](#openpose-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
|
|
@ -12,11 +13,10 @@
|
|||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Distributed Training](#distributed-training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
|
||||
# [Openpose Description](#contents)
|
||||
|
|
@ -294,7 +294,7 @@ Data storage method is the same as training
|
|||
| Model Version | openpose
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8
|
||||
| uploaded Date | 12/14/2020 (month/day/year)
|
||||
| MindSpore Version | 1.0.1-alpha
|
||||
| MindSpore Version | 1.0.1
|
||||
| Training Parameters | epoch=60(1pcs)/80(8pcs), steps=30k(1pcs)/5k(8pcs), batch_size=10, init_lr=0.0001
|
||||
| Optimizer | Adam(1pcs)/Momentum(8pcs)
|
||||
| Loss Function | MSE
|
||||
|
|
|
|||
|
|
@ -19,15 +19,12 @@
|
|||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [KingsCollege上的PoseNet](#KingsCollege上的PoseNet)
|
||||
- [StMarysChurch上的PoseNet](#StMarysChurch上的PoseNet)
|
||||
- [KingsCollege上的PoseNet](#kingscollege上的posenet)
|
||||
- [StMarysChurch上的PoseNet](#stmaryschurch上的posenet)
|
||||
- [推理性能](#推理性能)
|
||||
- [KingsCollege上的PoseNet](#KingsCollege上的PoseNet)
|
||||
- [StMarysChurch上的PoseNet](#StMarysChurch上的PoseNet)
|
||||
- [使用流程](#使用流程)
|
||||
- [推理](#推理)
|
||||
- [继续训练预训练模型](#继续训练预训练模型)
|
||||
- [迁移学习](#迁移学习)
|
||||
- [KingsCollege上的PoseNet](#kingscollege上的posenet-1)
|
||||
- [StMarysChurch上的PoseNet](#stmaryschurch上的posenet-1)
|
||||
- [迁移学习](#迁移学习)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
|
|
@ -291,7 +288,7 @@ PoseNet是剑桥大学提出的一种鲁棒、实时的6DOF（单目六自由度
|
|||
| -------------------------- | ----------------------------------------------------------- | ---------------------- |
|
||||
| 资源 | Ascend 910 ；CPU 2.60GHz，192核；内存：755G | NV SMX2 V100-32G |
|
||||
| 上传日期 | 2021-03-26 | 2021-05-20 |
|
||||
| MindSpore版本 | 1.1.1-alpha | 1.2.1-alpha |
|
||||
| MindSpore版本 | 1.1.1 | 1.2.1 |
|
||||
| 数据集 | KingsCollege | KingsCollege |
|
||||
| 训练参数 | max_steps=30000, batch_size=75, lr_init=0.001 | max_steps=30000, batch_size=75, lr_init=0.001 |
|
||||
| 优化器 | Adagrad | Adagrad |
|
||||
|
|
@ -311,7 +308,7 @@ PoseNet是剑桥大学提出的一种鲁棒、实时的6DOF（单目六自由度
|
|||
| -------------------------- | ----------------------------------------------------------- | ---------------------- |
|
||||
| 资源 | Ascend 910 ；CPU 2.60GHz，192核；内存：755G | NV SMX2 V100-32G |
|
||||
| 上传日期 | 2021-03-26 | 2021-05-20 |
|
||||
| MindSpore版本 | 1.1.1-alpha | 1.2.1-alpha |
|
||||
| MindSpore版本 | 1.1.1 | 1.2.1 |
|
||||
| 数据集 | StMarysChurch | StMarysChurch |
|
||||
| 训练参数 | max_steps=30000, batch_size=75, lr_init=0.001 | max_steps=30000, batch_size=75, lr_init=0.001 |
|
||||
| 优化器 | Adagrad | Adagrad |
|
||||
|
|
@ -333,7 +330,7 @@ PoseNet是剑桥大学提出的一种鲁棒、实时的6DOF（单目六自由度
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| 资源 | Ascend 910 | GPU |
|
||||
| 上传日期 | 2021-03-26 | 2021-05-20 |
|
||||
| MindSpore 版本 | 1.1.1-alpha | 1.2.1-alpha |
|
||||
| MindSpore 版本 | 1.1.1 | 1.2.1 |
|
||||
| 数据集 | KingsCollege | KingsCollege |
|
||||
| batch_size | 1 | 1 |
|
||||
| 输出 | 距离、角度 |距离、角度 |
|
||||
|
|
@ -346,7 +343,7 @@ PoseNet是剑桥大学提出的一种鲁棒、实时的6DOF（单目六自由度
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| 资源 | Ascend 910 | GPU |
|
||||
| 上传日期 | 2021-03-26 | 2021-05-20 |
|
||||
| MindSpore 版本 | 1.1.1-alpha | 1.2.1-alpha |
|
||||
| MindSpore 版本 | 1.1.1 | 1.2.1 |
|
||||
| 数据集 | StMarysChurch | StMarysChurch |
|
||||
| batch_size | 1 | 1 |
|
||||
| 输出 | 距离、角度 | 距离、角度 |
|
||||
|
|
|
|||
|
|
@ -1,6 +1,9 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [ResNet Description](#resnet-description)
|
||||
- [Description](#description)
|
||||
- [Paper](#paper)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
|
|
@ -11,15 +14,39 @@
|
|||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Usage](#usage)
|
||||
- [Running on Ascend](#running-on-ascend)
|
||||
- [Running on GPU](#running-on-gpu)
|
||||
- [Running parameter server mode training](#running-parameter-server-mode-training)
|
||||
- [Evaluation while training](#evaluation-while-training)
|
||||
- [Result](#result)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Usage](#usage-1)
|
||||
- [Running on Ascend](#running-on-ascend-1)
|
||||
- [Running on GPU](#running-on-gpu-1)
|
||||
- [Result](#result-1)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result-2)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [ResNet18 on CIFAR-10](#resnet18-on-cifar-10)
|
||||
- [ResNet18 on ImageNet2012](#resnet18-on-imagenet2012)
|
||||
- [ResNet50 on CIFAR-10](#resnet50-on-cifar-10)
|
||||
- [ResNet50 on ImageNet2012](#resnet50-on-imagenet2012)
|
||||
- [ResNet34 on ImageNet2012](#resnet34-on-imagenet2012)
|
||||
- [ResNet101 on ImageNet2012](#resnet101-on-imagenet2012)
|
||||
- [SE-ResNet50 on ImageNet2012](#se-resnet50-on-imagenet2012)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [ResNet18 on CIFAR-10](#resnet18-on-cifar-10-1)
|
||||
- [ResNet18 on ImageNet2012](#resnet18-on-imagenet2012-1)
|
||||
- [ResNet34 on ImageNet2012](#resnet34-on-imagenet2012-1)
|
||||
- [ResNet50 on CIFAR-10](#resnet50-on-cifar-10-1)
|
||||
- [ResNet50 on ImageNet2012](#resnet50-on-imagenet2012-1)
|
||||
- [ResNet101 on ImageNet2012](#resnet101-on-imagenet2012-1)
|
||||
- [SE-ResNet50 on ImageNet2012](#se-resnet50-on-imagenet2012-1)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -690,7 +717,7 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| Model Version | ResNet18 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 02/25/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.1.1-alpha |
|
||||
| MindSpore Version | 1.1.1 |
|
||||
| Dataset | CIFAR-10 |
|
||||
| Training Parameters | epoch=90, steps per epoch=195, batch_size = 32 |
|
||||
| Optimizer | Momentum |
|
||||
|
|
@ -710,7 +737,7 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| Model Version | ResNet18 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 02/25/2021 (month/day/year) ； |
|
||||
| MindSpore Version | 1.1.1-alpha |
|
||||
| MindSpore Version | 1.1.1 |
|
||||
| Dataset | ImageNet2012 |
|
||||
| Training Parameters | epoch=90, steps per epoch=626, batch_size = 256 |
|
||||
| Optimizer | Momentum |
|
||||
|
|
@ -729,8 +756,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| -------------------------- | -------------------------------------- |---------------------------------- |
|
||||
| Model Version | ResNet50-v1.5 |ResNet50-v1.5|
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | GPU(Tesla V100 SXM2)，CPU 2.1GHz 24cores，Memory 128G
|
||||
| uploaded Date | 04/01/2020 (month/day/year) | 08/01/2020 (month/day/year)
|
||||
| MindSpore Version | 0.1.0-alpha |0.6.0-alpha |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year)
|
||||
| MindSpore Version | 1.3.0 |1.3.0 |
|
||||
| Dataset | CIFAR-10 | CIFAR-10
|
||||
| Training Parameters | epoch=90, steps per epoch=195, batch_size = 32 |epoch=90, steps per epoch=195, batch_size = 32 |
|
||||
| Optimizer | Momentum |Momentum|
|
||||
|
|
@ -749,8 +776,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| -------------------------- | -------------------------------------- |---------------------------------- |
|
||||
| Model Version | ResNet50-v1.5 |ResNet50-v1.5|
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | GPU(Tesla V100 SXM2)，CPU 2.1GHz 24cores，Memory 128G
|
||||
| uploaded Date | 04/01/2020 (month/day/year) ； | 08/01/2020 (month/day/year)
|
||||
| MindSpore Version | 0.1.0-alpha |0.6.0-alpha |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) ； | 07/05/2021 (month/day/year)
|
||||
| MindSpore Version | 1.3.0 |1.3.0 |
|
||||
| Dataset | ImageNet2012 | ImageNet2012|
|
||||
| Training Parameters | epoch=90, steps per epoch=626, batch_size = 256 |epoch=90, steps per epoch=626, batch_size = 256 |
|
||||
| Optimizer | Momentum |Momentum|
|
||||
|
|
@ -769,8 +796,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| -------------------------- | -------------------------------------- |---------------------------------- |
|
||||
| Model Version | ResNet50-v1.5 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 04/01/2020 (month/day/year) ； |
|
||||
| MindSpore Version | 0.1.0-alpha |
|
||||
| uploaded Date | 07/05/2020 (month/day/year) ； |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | ImageNet2012 |
|
||||
| Training Parameters | epoch=90, steps per epoch=626, batch_size = 256 |
|
||||
| Optimizer | Momentum |
|
||||
|
|
@ -789,8 +816,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| -------------------------- | -------------------------------------- |---------------------------------- |
|
||||
| Model Version | ResNet101 |ResNet101|
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | GPU(Tesla V100 SXM2)，CPU 2.1GHz 24cores，Memory 128G
|
||||
| uploaded Date | 04/01/2020 (month/day/year) | 08/01/2020 (month/day/year)
|
||||
| MindSpore Version | 0.1.0-alpha |0.6.0-alpha |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year)
|
||||
| MindSpore Version | 1.3.0 |1.3.0 |
|
||||
| Dataset | ImageNet2012 | ImageNet2012|
|
||||
| Training Parameters | epoch=120, steps per epoch=5004, batch_size = 32 |epoch=120, steps per epoch=5004, batch_size = 32 |
|
||||
| Optimizer | Momentum |Momentum|
|
||||
|
|
@ -809,8 +836,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| -------------------------- | ------------------------------------------------------------------------ |
|
||||
| Model Version | SE-ResNet50 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 08/16/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.7.0-alpha |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | ImageNet2012 |
|
||||
| Training Parameters | epoch=24, steps per epoch=5004, batch_size = 32 |
|
||||
| Optimizer | Momentum |
|
||||
|
|
@ -832,7 +859,7 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| Model Version | ResNet18 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 02/25/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.1.1-alpha |
|
||||
| MindSpore Version | 1.1.1 |
|
||||
| Dataset | CIFAR-10 |
|
||||
| batch_size | 32 |
|
||||
| outputs | probability |
|
||||
|
|
@ -846,7 +873,7 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| Model Version | ResNet18 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 02/25/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.1.1-alpha |
|
||||
| MindSpore Version | 1.1.1 |
|
||||
| Dataset | ImageNet2012 |
|
||||
| batch_size | 256 |
|
||||
| outputs | probability |
|
||||
|
|
@ -860,7 +887,7 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| Model Version | ResNet18 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 02/25/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.1.1-alpha |
|
||||
| MindSpore Version | 1.1.1 |
|
||||
| Dataset | ImageNet2012 |
|
||||
| batch_size | 256 |
|
||||
| outputs | probability |
|
||||
|
|
@ -873,8 +900,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| Model Version | ResNet50-v1.5 | ResNet50-v1.5 |
|
||||
| Resource | Ascend 910; OS Euler2.8 | GPU |
|
||||
| Uploaded Date | 04/01/2020 (month/day/year) | 08/01/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.1.0-alpha | 0.6.0-alpha |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | CIFAR-10 | CIFAR-10 |
|
||||
| batch_size | 32 | 32 |
|
||||
| outputs | probability | probability |
|
||||
|
|
@ -887,8 +914,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| Model Version | ResNet50-v1.5 | ResNet50-v1.5 |
|
||||
| Resource | Ascend 910; OS Euler2.8 | GPU |
|
||||
| Uploaded Date | 04/01/2020 (month/day/year) | 08/01/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.1.0-alpha | 0.6.0-alpha |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet2012 | ImageNet2012 |
|
||||
| batch_size | 256 | 256 |
|
||||
| outputs | probability | probability |
|
||||
|
|
@ -900,9 +927,9 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| Parameters | Ascend | GPU |
|
||||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| Model Version | ResNet101 | ResNet101 |
|
||||
| Resource | Ascend 910; OS Euler2.8 | GPU |
|
||||
| Uploaded Date | 04/01/2020 (month/day/year) | 08/01/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.1.0-alpha | 0.6.0-alpha |
|
||||
| Resource | Ascend 910; OS Euler2.8 | GPU |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet2012 | ImageNet2012 |
|
||||
| batch_size | 32 | 32 |
|
||||
| outputs | probability | probability |
|
||||
|
|
@ -915,8 +942,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| ------------------- | --------------------------- |
|
||||
| Model Version | SE-ResNet50 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 08/16/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.7.0-alpha |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | ImageNet2012 |
|
||||
| batch_size | 32 |
|
||||
| outputs | probability |
|
||||
|
|
|
|||
|
|
@ -2,7 +2,10 @@
|
|||
|
||||
<!-- TOC -->
|
||||
|
||||
- [ResNet描述](#ResNet描述)
|
||||
- [目录](#目录)
|
||||
- [ResNet描述](#resnet描述)
|
||||
- [概述](#概述)
|
||||
- [论文](#论文)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [特性](#特性)
|
||||
|
|
@ -13,16 +16,33 @@
|
|||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [用法](#用法)
|
||||
- [Ascend处理器环境运行](#ascend处理器环境运行)
|
||||
- [GPU处理器环境运行](#gpu处理器环境运行)
|
||||
- [运行参数服务器模式训练](#运行参数服务器模式训练)
|
||||
- [训练时推理](#训练时推理)
|
||||
- [结果](#结果)
|
||||
- [评估过程](#评估过程)
|
||||
- [用法](#用法-1)
|
||||
- [Ascend处理器环境运行](#ascend处理器环境运行-1)
|
||||
- [GPU处理器环境运行](#gpu处理器环境运行-1)
|
||||
- [结果](#结果-1)
|
||||
- [推理过程](#推理过程)
|
||||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [结果](#结果-2)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [CIFAR-10上的ResNet18](#cifar-10上的resnet18)
|
||||
- [ImageNet2012上的ResNet18](#imagenet2012上的resnet18)
|
||||
- [CIFAR-10上的ResNet50](#cifar-10上的resnet50)
|
||||
- [ImageNet2012上的ResNet50](#imagenet2012上的resnet50)
|
||||
- [ImageNet2012上的ResNet34](#imagenet2012上的resnet34)
|
||||
- [ImageNet2012上的ResNet101](#imagenet2012上的resnet101)
|
||||
- [ImageNet2012上的SE-ResNet50](#imagenet2012上的se-resnet50)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#ModelZoo主页)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -658,7 +678,7 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| 模型版本 | ResNet18 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |
|
||||
| 上传日期 | 2021-02-25 |
|
||||
| MindSpore版本 | 1.1.1-alpha |
|
||||
| MindSpore版本 | 1.1.1 |
|
||||
| 数据集 | CIFAR-10 |
|
||||
| 训练参数 | epoch=90, steps per epoch=195, batch_size = 32 |
|
||||
| 优化器 | Momentum |
|
||||
|
|
@ -678,7 +698,7 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| 模型版本 | ResNet18 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-04-01 ; |
|
||||
| MindSpore版本 | 1.1.1-alpha |
|
||||
| MindSpore版本 | 1.1.1 |
|
||||
| 数据集 | ImageNet2012 |
|
||||
| 训练参数 | epoch=90, steps per epoch=626, batch_size = 256 |
|
||||
| 优化器 | Momentum |
|
||||
|
|
@ -697,8 +717,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| -------------------------- | -------------------------------------- |---------------------------------- |
|
||||
| 模型版本 | ResNet50-v1.5 |ResNet50-v1.5|
|
||||
| 资源 |Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 | GPU(Tesla V100 SXM2)；CPU：2.1GHz，24核；内存：128G
|
||||
| 上传日期 | 2020-04-01 | 2020-08-01
|
||||
| MindSpore版本 | 0.1.0-alpha |0.6.0-alpha |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05
|
||||
| MindSpore版本 | 1.3.0 |1.3.0 |
|
||||
| 数据集 | CIFAR-10 | CIFAR-10
|
||||
| 训练参数 | epoch=90, steps per epoch=195, batch_size = 32 |epoch=90, steps per epoch=195, batch_size = 32 |
|
||||
| 优化器 | Momentum |Momentum|
|
||||
|
|
@ -717,8 +737,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| -------------------------- | -------------------------------------- |---------------------------------- |
|
||||
| 模型版本 | ResNet50-v1.5 |ResNet50-v1.5|
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 | GPU(Tesla V100 SXM2)；CPU：2.1GHz，24核；内存：128G
|
||||
| 上传日期 | 2020-04-01 ; | 2020-08-01
|
||||
| MindSpore版本 | 0.1.0-alpha |0.6.0-alpha |
|
||||
| 上传日期 | 2021-07-05 ; | 2021-07-05
|
||||
| MindSpore版本 | 1.3.0 |1.3.0 |
|
||||
| 数据集 | ImageNet2012 | ImageNet2012|
|
||||
| 训练参数 | epoch=90, steps per epoch=626, batch_size = 256 |epoch=90, steps per epoch=5004, batch_size = 32 |
|
||||
| 优化器 | Momentum |Momentum|
|
||||
|
|
@ -738,7 +758,7 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| 模型版本 | ResNet34 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |
|
||||
| 上传日期 | 2021-05-08 ; |
|
||||
| MindSpore版本 | 1.1.1-alpha |
|
||||
| MindSpore版本 | 1.1.1 |
|
||||
| 数据集 | ImageNet2012 |
|
||||
| 训练参数 | epoch=90, steps per epoch=625, batch_size = 256 |
|
||||
| 优化器 | Momentum |
|
||||
|
|
@ -757,8 +777,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| -------------------------- | -------------------------------------- |---------------------------------- |
|
||||
| 模型版本 | ResNet101 |ResNet101|
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 | GPU(Tesla V100 SXM2)；CPU：2.1GHz，24核；内存：128G
|
||||
| 上传日期 | 2020-04-01 ; | 2020-08-01
|
||||
| MindSpore版本 | 0.1.0-alpha |0.6.0-alpha |
|
||||
| 上传日期 | 2021-07-05 ; | 2021-07-05
|
||||
| MindSpore版本 | 1.3.0 |1.3.0 |
|
||||
| 数据集 | ImageNet2012 | ImageNet2012|
|
||||
| 训练参数 | epoch=120, steps per epoch=5004, batch_size = 32 |epoch=120, steps per epoch=5004, batch_size = 32 |
|
||||
| 优化器 | Momentum |Momentum|
|
||||
|
|
@ -777,8 +797,8 @@ Total data: 50000, top1 accuracy: 0.76844, top5 accuracy: 0.93522.
|
|||
| -------------------------- | ------------------------------------------------------------------------ |
|
||||
| 模型版本 | SE-ResNet50 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-08-16 ； |
|
||||
| MindSpore版本 | 0.7.0-alpha |
|
||||
| 上传日期 | 2021-07-05 ； |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | ImageNet2012 |
|
||||
| 训练参数 | epoch=24, steps per epoch=5004, batch_size = 32 |
|
||||
| 优化器 | Momentum |
|
||||
|
|
|
|||
|
|
@ -1,17 +1,26 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [ResNet50 Description](#resnet50-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script description](#script-description)
|
||||
- [Script and sample code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Model Description](#model-description)
|
||||
- [Training process](#training-process)
|
||||
- [Usage](#usage)
|
||||
- [Launch](#launch)
|
||||
- [Result](#result)
|
||||
- [Evaluation process](#evaluation-process)
|
||||
- [Usage](#usage-1)
|
||||
- [Launch](#launch-1)
|
||||
- [Result](#result-1)
|
||||
- [Model Export](#model-export)
|
||||
- [Ascend 310 inference](#ascend-310-inference)
|
||||
- [Model description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
|
|
@ -213,8 +222,8 @@ You can view the results through the file "acc.log". The accuracy of the test da
|
|||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| Model Version | ResNet50 V1.5 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 56cores; Memory 314G; OS Euler2.8 |
|
||||
| uploaded Date | 06/06/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.3.0-alpha |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | ImageNet |
|
||||
| Training Parameters | epoch=30(with pretrained) or 120, steps per epoch=5004, batch_size=32 |
|
||||
| Optimizer | Momentum |
|
||||
|
|
@ -233,8 +242,8 @@ You can view the results through the file "acc.log". The accuracy of the test da
|
|||
| ------------------- | --------------------------- |
|
||||
| Model Version | ResNet50 V1.5 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 56cores; Memory 314G; OS Euler2.8 |
|
||||
| Uploaded Date | 06/06/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.3.0-alpha |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | ImageNet |
|
||||
| batch_size | 32 |
|
||||
| outputs | probability |
|
||||
|
|
|
|||
|
|
@ -7,25 +7,27 @@
|
|||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [特性](#特性)
|
||||
- [混合精度](#混合精度)
|
||||
- [混合精度](#混合精度)
|
||||
- [环境要求](#环境要求)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本和样例代码](#脚本和样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [用法](#用法)
|
||||
- [启动](#启动)
|
||||
- [结果](#结果)
|
||||
- [评估过程](#评估过程)
|
||||
- [用法](#用法-1)
|
||||
- [启动](#启动-1)
|
||||
- [结果](#结果-1)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本和样例代码](#脚本和样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [用法](#用法)
|
||||
- [启动](#启动)
|
||||
- [结果](#结果)
|
||||
- [评估过程](#评估过程)
|
||||
- [用法](#用法-1)
|
||||
- [启动](#启动-1)
|
||||
- [结果](#结果-1)
|
||||
- [模型导出](#模型导出)
|
||||
- [Ascend 310 推理](#ascend-310-推理)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -221,8 +223,8 @@ bash run_infer_310.sh [AIR_PATH] [DATA_PATH] [LABEL_PATH] [DEVICE_ID]
|
|||
| -------------------------- | ---------------------------------------------------------- |
|
||||
| 模型版本 | V1 |
|
||||
| 资源 | Ascend 910； CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-06 |
|
||||
| MindSpore版本 | 0.3.0 |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | ImageNet |
|
||||
| 训练参数 | src/config.py |
|
||||
| 优化器 | Momentum |
|
||||
|
|
@ -241,8 +243,8 @@ bash run_infer_310.sh [AIR_PATH] [DATA_PATH] [LABEL_PATH] [DEVICE_ID]
|
|||
| -------------------------- | ----------------------------- |
|
||||
| 模型版本 | V1 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-06 |
|
||||
| MindSpore版本 | 0.3.0 |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | ImageNet, 1.2W |
|
||||
| batch_size | 130（8卡） |
|
||||
| 输出 | 概率 |
|
||||
|
|
|
|||
|
|
@ -1,21 +1,26 @@
|
|||
# ResNet-50-THOR Example
|
||||
|
||||
- [Description](#Description)
|
||||
- [Model Architecture](#Model-Architecture)
|
||||
- [Dataset](#Dataset)
|
||||
- [Features](#Features)
|
||||
- [Environment Requirements](#Environment-Requirements)
|
||||
- [Quick Start](#Quick-Start)
|
||||
- [Script Description](#Script-Description)
|
||||
- [Script and Sample Code](#Script-Code-Structure)
|
||||
- [Script Parameters](#Script-Parameters)
|
||||
- [Training Process](#Training-Process)
|
||||
- [Evaluation Process](#Evaluation-Process)
|
||||
- [Model Description](#Model-Description)
|
||||
- [Evaluation Performance](#Evaluation-Performance)
|
||||
- [Inference Performance](#Inference-Performance)
|
||||
- [Description of Random Situation](#Description-of-Random-Situation)
|
||||
- [ModelZoo Homepage](#ModelZoo-Homepage)
|
||||
- [ResNet-50-THOR Example](#resnet-50-thor-example)
|
||||
- [Description](#description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script Code Structure](#script-code-structure)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Ascend 910](#ascend-910)
|
||||
- [GPU](#gpu)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Ascend 910](#ascend-910-1)
|
||||
- [GPU](#gpu-1)
|
||||
- [Model Description](#model-description)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo HomePage](#modelzoo-homepage)
|
||||
|
||||
## Description
|
||||
|
||||
|
|
@ -259,8 +264,8 @@ Inference result will be stored in the example path, whose folder name is "eval"
|
|||
| -------------------------- | -------------------------------------- |---------------------------------- |
|
||||
| Model Version | ResNet50-v1.5 |ResNet50-v1.5|
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | GPU(Tesla V100 SXM2)，CPU 2.1GHz 24cores，Memory 128G
|
||||
| uploaded Date | 06/01/2020 (month/day/year) | 09/23/2020(month/day/year)
|
||||
| MindSpore Version | 0.3.0-alpha | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021(month/day/year)
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet2012 | ImageNet2012|
|
||||
| Training Parameters | epoch=45, steps per epoch=5004, batch_size = 32 |epoch=40, steps per epoch=5004, batch_size = 32 |
|
||||
| Optimizer | THOR |THOR|
|
||||
|
|
@ -279,8 +284,8 @@ Inference result will be stored in the example path, whose folder name is "eval"
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| Model Version | ResNet50-v1.5 | ResNet50-v1.5 |
|
||||
| Resource | Ascend 910; OS Euler2.8 | GPU |
|
||||
| Uploaded Date | 06/01/2020 (month/day/year) | 09/23/2020(month/day/year) |
|
||||
| MindSpore Version | 0.3.0-alpha | 1.0.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021(month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet2012 | ImageNet2012 |
|
||||
| batch_size | 32 | 32 |
|
||||
| outputs | probability | probability |
|
||||
|
|
|
|||
|
|
@ -13,12 +13,16 @@
|
|||
- [脚本代码结构](#脚本代码结构)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [Ascend 910](#ascend-910)
|
||||
- [GPU](#gpu)
|
||||
- [推理过程](#推理过程)
|
||||
- [Ascend 910](#ascend-910-1)
|
||||
- [GPU](#gpu-1)
|
||||
- [模型描述](#模型描述)
|
||||
- [训练性能](#训练性能)
|
||||
- [推理性能](#推理性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo首页](#ModelZoo首页)
|
||||
- [ModelZoo首页](#modelzoo首页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -265,8 +269,8 @@ epoch： 36 step: 5004，loss is 1.645802
|
|||
| -------------------------- | -------------------------------------- | ---------------------------------- |
|
||||
| 模型版本 | ResNet50-v1.5 | ResNet50-v1.5 |
|
||||
| 资源 |Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 | GPU(Tesla V100 SXM2)-CPU 2.1GHz 24核-内存128G |
|
||||
| 上传日期 | 2020-06-01 | 2020-09-23 |
|
||||
| MindSpore版本 | 0.3.0-alpha | 1.0.0|
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0|
|
||||
| 数据集 | ImageNet2012 | ImageNet2012 |
|
||||
| 训练参数 | epoch=45, steps per epoch=5004, batch_size = 32 |epoch=40, steps per epoch=5004, batch_size = 32 |
|
||||
| 优化器 |THOR|THOR |
|
||||
|
|
@ -285,8 +289,8 @@ epoch： 36 step: 5004，loss is 1.645802
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| 模型版本 | ResNet50-v1.5 | ResNet50-v1.5 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | GPU |
|
||||
| 上传日期 | 2020-06-01 | 2020-09-23 |
|
||||
| MindSpore版本 | 0.3.0-alpha | 1.0.0 |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | ImageNet2012 | ImageNet2012 |
|
||||
| 批大小 | 32 | 32 |
|
||||
| 输出 | 概率 | 概率 |
|
||||
|
|
|
|||
|
|
@ -1,23 +1,30 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [ResNeXt Description](#resnext-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Model architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script description](#script-description)
|
||||
- [Script and sample code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Usage](#usage)
|
||||
- [Launch](#launch)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Usage](#usage-1)
|
||||
- [Launch](#launch-1)
|
||||
- [Result](#result)
|
||||
- [Model Export](#model-export)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#evaluation-performance)
|
||||
- [Inference Performance](#evaluation-performance)
|
||||
- [Usage](#usage-2)
|
||||
- [result](#result-1)
|
||||
- [Model description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -311,8 +318,8 @@ Total data:50000, top1 accuracy:0.79858, top5 accuracy:0.94716
|
|||
| Parameters | ResNeXt50 | |
|
||||
| -------------------------- | ---------------------------------------------------------- | ------------------------- |
|
||||
| Resource | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 | NV SMX2 V100-32G |
|
||||
| uploaded Date | 06/30/2020 | 07/23/2020 |
|
||||
| MindSpore Version | 0.5.0 | 0.6.0 |
|
||||
| uploaded Date | 07/05/2021 | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet | ImageNet |
|
||||
| Training Parameters | default_config.yaml | default_config.yaml |
|
||||
| Optimizer | Momentum | Momentum |
|
||||
|
|
@ -339,8 +346,8 @@ Total data:50000, top1 accuracy:0.79858, top5 accuracy:0.94716
|
|||
| Parameters | ResNeXt50 | | |
|
||||
| -------------------------- | ----------------------------- | ------------------------- | -------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 | NV SMX2 V100-32G | Ascend 310 |
|
||||
| uploaded Date | 06/30/2020 | 07/23/2020 | 07/23/2020 |
|
||||
| MindSpore Version | 0.5.0 | 0.6.0 | 0.6.0 |
|
||||
| uploaded Date | 07/05/2021 | 07/05/2021 | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet, 1.2W | ImageNet, 1.2W | ImageNet, 1.2W |
|
||||
| batch_size | 1 | 1 | 1 |
|
||||
| outputs | probability | probability | probability |
|
||||
|
|
|
|||
|
|
@ -20,7 +20,7 @@
|
|||
- [模型导出](#模型导出)
|
||||
- [推理过程](#推理过程)
|
||||
- [用法](#用法-2)
|
||||
- [结果](#结果-2)
|
||||
- [结果](#结果-1)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
|
|
@ -323,8 +323,8 @@ Total data:50000, top1 accuracy:0.79858, top5 accuracy:0.94716
|
|||
| 参数 | ResNeXt50 | |
|
||||
| -------------------------- | ---------------------------------------------------------- | ------------------------- |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755GB；系统 Euler2.8 | NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-6-30 | 2020-7-23 |
|
||||
| MindSpore版本 | 0.5.0 | 0.6.0 |
|
||||
| 上传日期 | 2021-7-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | ImageNet | ImageNet |
|
||||
| 训练参数 | src/config.py | src/config.py |
|
||||
| 优化器 | Momentum | Momentum |
|
||||
|
|
@ -339,8 +339,8 @@ Total data:50000, top1 accuracy:0.79858, top5 accuracy:0.94716
|
|||
| 参数 |ResNeXt50 | | |
|
||||
| -------------------------- | ----------------------------- | ------------------------- | -------------------- |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | NV SMX2 V100-32G | Ascend 310 |
|
||||
| 上传日期 | 2020-6-30 | 2020-7-23 | 2020-7-23 |
|
||||
| MindSpore版本 | 0.5.0 | 0.6.0 | 0.6.0 |
|
||||
| 上传日期 | 2021-7-05 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | ImageNet， 1.2万 | ImageNet， 1.2万 | ImageNet， 1.2万 |
|
||||
| batch_size | 1 | 1 | 1 |
|
||||
| 输出 | 概率 | 概率 | 概率 |
|
||||
|
|
@ -349,7 +349,7 @@ Total data:50000, top1 accuracy:0.79858, top5 accuracy:0.94716
|
|||
| 参数 | ResNeXt101 |
|
||||
| ------------------- | --------------------------- |
|
||||
| 资源 | Ascend 310; OS Euler2.8 |
|
||||
| 上传日期 | 22/06/2021 (month/day/year) |
|
||||
| 上传日期 | 06/22/2021 (month/day/year) |
|
||||
| MindSpore版本 | 1.2.0 |
|
||||
| 数据集 | ImageNet |
|
||||
| batch_size | 1 |
|
||||
|
|
|
|||
|
|
@ -14,6 +14,7 @@
|
|||
- [Training Process](#training-process)
|
||||
- [Training on Ascend](#training-on-ascend)
|
||||
- [Training on GPU](#training-on-gpu)
|
||||
- [Evaluation while training](#evaluation-while-training)
|
||||
- [Transfer Training](#transfer-training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation on Ascend](#evaluation-on-ascend)
|
||||
|
|
@ -545,8 +546,8 @@ mAP: 0.33880018942412393
|
|||
| ------------------- | ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------- |
|
||||
| Model Version | SSD V1 | SSD V1 | SSD-Mobilenet-V1-Fpn |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | NV SMX2 V100-16G | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 09/15/2020 (month/day/year) | 09/24/2020 (month/day/year) | 01/13/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 | 1.0.0 | 1.1.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) | 09/24/2020 (month/day/year) | 01/13/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.0.0 | 1.1.0 |
|
||||
| Dataset | COCO2017 | COCO2017 | COCO2017 |
|
||||
| Training Parameters | epoch = 500, batch_size = 32 | epoch = 800, batch_size = 32 | epoch = 60, batch_size = 32 |
|
||||
| Optimizer | Momentum | Momentum | Momentum |
|
||||
|
|
@ -562,8 +563,8 @@ mAP: 0.33880018942412393
|
|||
| ------------------- | --------------------------- | --------------------------- | --------------------------- |
|
||||
| Model Version | SSD V1 | SSD V1 | SSD-Mobilenet-V1-Fpn |
|
||||
| Resource | Ascend 910; OS Euler2.8 | GPU |Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 09/15/2020 (month/day/year) | 09/24/2020 (month/day/year) | 09/24/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 | 1.0.0 | 1.1.0 |
|
||||
| Uploaded Date | 07/05/2020 (month/day/year) | 09/24/2020 (month/day/year) | 09/24/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.0.0 | 1.1.0 |
|
||||
| Dataset | COCO2017 | COCO2017 | COCO2017 |
|
||||
| batch_size | 1 | 1 | 1 |
|
||||
| outputs | mAP | mAP | mAP |
|
||||
|
|
|
|||
|
|
@ -468,8 +468,8 @@ mAP: 0.33880018942412393
|
|||
| -------------------------- | -------------------------------------------------------------| -------------------------------------------------------------|
|
||||
| 模型版本 | SSD V1 | SSD V1 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755GB；系统 Euler2.8 | NV SMX2 V100-16G |
|
||||
| 上传日期 | 2020-06-01 | 2020-09-24 |
|
||||
| MindSpore版本 | 0.3.0-alpha | 1.0.0 |
|
||||
| 上传日期 | 2021-07-05 | 2020-09-24 |
|
||||
| MindSpore版本 | 1.3.0 | 1.0.0 |
|
||||
| 数据集 | COCO2017 | COCO2017 |
|
||||
| 训练参数 | epoch = 500, batch_size = 32 | epoch = 800, batch_size = 32 |
|
||||
| 优化器 | Momentum | Momentum |
|
||||
|
|
@ -485,8 +485,8 @@ mAP: 0.33880018942412393
|
|||
| ------------------- | ----------------------------| ----------------------------|
|
||||
| 模型版本 | SSD V1 | SSD V1 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | GPU |
|
||||
| 上传日期 | 2020-06-01 | 2020-09-24 |
|
||||
| MindSpore版本 | 0.3.0-alpha | 1.0.0 |
|
||||
| 上传日期 | 2021-07-05 | 2020-09-24 |
|
||||
| MindSpore版本 | 1.3.0 | 1.0.0 |
|
||||
| 数据集 | COCO2017 | COCO2017 |
|
||||
| batch_size | 1 | 1 |
|
||||
| 输出 | mAP | mAP |
|
||||
|
|
|
|||
|
|
@ -1,30 +1,38 @@
|
|||
# Contents
|
||||
|
||||
- [VGG Description](#vgg-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Parameter configuration](#parameter-configuration)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Contents](#contents)
|
||||
- [VGG Description](#vgg-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Dataset used: CIFAR-10](#dataset-used-cifar-10)
|
||||
- [Dataset used: ImageNet2012](#dataset-used-imagenet2012)
|
||||
- [Dataset organize way](#dataset-organize-way)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training](#training)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Parameter configuration](#parameter-configuration)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training-1)
|
||||
- [Run vgg16 on Ascend](#run-vgg16-on-ascend)
|
||||
- [Run vgg16 on GPU](#run-vgg16-on-gpu)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation-1)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## [VGG Description](#contents)
|
||||
|
||||
|
|
@ -532,8 +540,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| -------------------------- | ---------------------------------------------- |------------------------------------|
|
||||
| Model Version | VGG16 | VGG16 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |NV SMX2 V100-32G |
|
||||
| uploaded Date | 10/28/2020 | 10/28/2020 |
|
||||
| MindSpore Version | 1.0.0 | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | CIFAR-10 |ImageNet2012 |
|
||||
| Training Parameters | epoch=70, steps=781, batch_size = 64, lr=0.1 |epoch=150, steps=40036, batch_size = 32, lr=0.1 |
|
||||
| Optimizer | Momentum |Momentum |
|
||||
|
|
@ -551,8 +559,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| ------------------- | --------------------------- |---------------------
|
||||
| Model Version | VGG16 | VGG16 |
|
||||
| Resource | Ascend 910; OS Euler2.8 | GPU |
|
||||
| Uploaded Date | 10/28/2020 | 10/28/2020 |
|
||||
| MindSpore Version | 1.0.0 | 1.0.0 |
|
||||
| Uploaded Date | 07/05/2021 | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | CIFAR-10, 10,000 images |ImageNet2012, 5000 images |
|
||||
| batch_size | 64 | 32 |
|
||||
| outputs | probability | probability |
|
||||
|
|
|
|||
|
|
@ -3,31 +3,38 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [VGG描述](#vgg描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [特性](#特性)
|
||||
- [混合精度](#混合精度)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练](#训练)
|
||||
- [评估](#评估)
|
||||
- [参数配置](#参数配置)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练-1)
|
||||
- [Ascend处理器环境运行VGG16](#ascend处理器环境运行vgg16)
|
||||
- [GPU处理器环境运行VGG16](#gpu处理器环境运行vgg16)
|
||||
- [评估过程](#评估过程)
|
||||
- [评估](#评估-1)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [VGG描述](#vgg描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [使用的数据集：CIFAR-10](#使用的数据集cifar-10)
|
||||
- [使用的数据集：ImageNet2012](#使用的数据集imagenet2012)
|
||||
- [数据集组织方式](#数据集组织方式)
|
||||
- [特性](#特性)
|
||||
- [混合精度](#混合精度)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练](#训练)
|
||||
- [评估](#评估)
|
||||
- [参数配置](#参数配置)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练-1)
|
||||
- [Ascend处理器环境运行VGG16](#ascend处理器环境运行vgg16)
|
||||
- [GPU处理器环境运行VGG16](#gpu处理器环境运行vgg16)
|
||||
- [评估过程](#评估过程)
|
||||
- [评估](#评估-1)
|
||||
- [推理过程](#推理过程)
|
||||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -536,8 +543,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATASET_NAME] [DATASET_PATH] [NEED_PREPROCE
|
|||
| -------------------------- | ---------------------------------------------- |------------------------------------|
|
||||
| 模型版本 | VGG16 | VGG16 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755GB；系统 Euler2.8 |NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-08-20 | 2020-08-20 |
|
||||
| MindSpore版本 | 0.5.0-alpha |0.5.0-alpha |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |1.3.0 |
|
||||
| 数据集 | CIFAR-10 |ImageNet2012 |
|
||||
| 训练参数 | epoch=70, steps=781, batch_size = 64, lr=0.1 |epoch=150, steps=40036, batch_size = 32, lr=0.1 |
|
||||
| 优化器 | Momentum | Momentum |
|
||||
|
|
@ -555,8 +562,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATASET_NAME] [DATASET_PATH] [NEED_PREPROCE
|
|||
| ------------------- | --------------------------- |---------------------
|
||||
| 模型版本 | VGG16 | VGG16 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | GPU |
|
||||
| 上传日期 | 2020-08-20 | 2020-08-20 |
|
||||
| MindSpore版本 | 0.5.0-alpha |0.5.0-alpha |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |1.3.0 |
|
||||
| 数据集 | CIFAR-10，10000张图像 | ImageNet2012，5000张图像 |
|
||||
| batch_size | 64 | 32 |
|
||||
| 输出 | 概率 | 概率 |
|
||||
|
|
|
|||
|
|
@ -1,32 +1,33 @@
|
|||
# Contents
|
||||
|
||||
- [WarpCTC Description](#warpctc-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Script Parameters](#training-script-parameters)
|
||||
- [Parameters Configuration](#parameters-configuration)
|
||||
- [Contents](#contents)
|
||||
- [WarpCTC Description](#warpctc-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Script Parameters](#training-script-parameters)
|
||||
- [Parameters Configuration](#parameters-configuration)
|
||||
- [Dataset Preparation](#dataset-preparation)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Distributed Training](#distributed-training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Distributed Training](#distributed-training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## [WarpCTC Description](#contents)
|
||||
|
||||
|
|
@ -348,8 +349,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| -------------------------- | --------------------------------------------- |---------------------------------- |
|
||||
| Model Version | v1.0 | v1.0 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | GPU(Tesla V100 SXM2)，CPU 2.1GHz 24cores，Memory 128G /
|
||||
| uploaded Date | 07/01/2020 (month/day/year) | 08/01/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.5.0-alpha | 0.6.0-alpha |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | Captcha | Captcha |
|
||||
| Training Parameters | epoch=30, steps per epoch=98, batch_size = 64 | epoch=30, steps per epoch=98, batch_size = 64 |
|
||||
| Optimizer | SGD | SGD |
|
||||
|
|
@ -368,8 +369,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| ------------------- | --------------------------- |
|
||||
| Model Version | V1.0 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 08/01/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.6.0-alpha |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | Captcha |
|
||||
| batch_size | 64 |
|
||||
| outputs | ACC |
|
||||
|
|
|
|||
|
|
@ -3,15 +3,15 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [WarpCTC描述](#warpctc描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练脚本参数](#训练脚本参数)
|
||||
- [WarpCTC描述](#warpctc描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练脚本参数](#训练脚本参数)
|
||||
- [参数配置](#参数配置)
|
||||
- [数据集准备](#数据集准备)
|
||||
- [训练过程](#训练过程)
|
||||
|
|
@ -23,13 +23,13 @@
|
|||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [推理性能](#推理性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -352,8 +352,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATA_PATH] [DEVICE_ID]
|
|||
| -------------------------- | --------------------------------------------- |---------------------------------- |
|
||||
| 模型版本 | v1.0 | v1.0 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 | GPU(Tesla V100 SXM2)，CPU 2.1GHz 24核，内存： 128G
|
||||
| 上传日期 | 2020-07-01 | 2020-08-01 |
|
||||
| MindSpore版本 | 0.5.0-alpha | 0.6.0-alpha |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | Captcha | Captcha |
|
||||
| 训练参数 | epoch=30, steps per epoch=98, batch_size = 64 | epoch=30, steps per epoch=98, batch_size = 64 |
|
||||
| 优化器 | SGD | SGD |
|
||||
|
|
@ -372,8 +372,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATA_PATH] [DEVICE_ID]
|
|||
| ------------------- | --------------------------- |
|
||||
| 模型版本 | V1.0 |
|
||||
| 资源 |Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-08-01 |
|
||||
| MindSpore版本 | 0.6.0-alpha |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | Captcha |
|
||||
| batch_size | 64 |
|
||||
| 输出 | ACC |
|
||||
|
|
|
|||
|
|
@ -3,25 +3,27 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [YOLOv3-DarkNet53-Quant描述](#yolov3-darknet53-quant描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [Ascend上训练](#ascend上训练)
|
||||
- [分布式训练](#分布式训练)
|
||||
- [评估过程](#评估过程)
|
||||
- [Ascend评估](#ascend评估)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [推理性能](#推理性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [YOLOv3-DarkNet53-Quant描述](#yolov3-darknet53-quant描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [Ascend上训练](#ascend上训练)
|
||||
- [分布式训练](#分布式训练)
|
||||
- [评估过程](#评估过程)
|
||||
- [Ascend评估](#ascend评估)
|
||||
- [模型导出](#模型导出)
|
||||
- [Ascend 310 推理](#ascend-310-推理)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [推理性能](#推理性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
|
|||
|
|
@ -1,5 +1,6 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [YOLOv3_ResNet18 Description](#yolov3_resnet18-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
|
|
@ -9,15 +10,17 @@
|
|||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Training on Ascend](#training-on-ascend)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Evaluation on Ascend](#evaluation-on-ascend)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Usage](#usage)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -365,8 +368,8 @@ Inference result is saved in current path, you can find result in acc.log file.
|
|||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| Model Version | YOLOv3_Resnet18 V1 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 09/15/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | COCO2017 |
|
||||
| Training Parameters | epoch = 160, batch_size = 32, lr = 0.005 |
|
||||
| Optimizer | Adam |
|
||||
|
|
@ -383,8 +386,8 @@ Inference result is saved in current path, you can find result in acc.log file.
|
|||
| ------------------- | ----------------------------------------------- |
|
||||
| Model Version | YOLOv3_Resnet18 V1 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 09/15/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | COCO2017 |
|
||||
| batch_size | 1 |
|
||||
| outputs | presion and recall |
|
||||
|
|
|
|||
|
|
@ -17,8 +17,8 @@
|
|||
- [Ascend评估](#ascend评估)
|
||||
- [导出mindir模型](#导出mindir模型)
|
||||
- [推理过程](#推理过程)
|
||||
- [用法](#用法-2)
|
||||
- [结果](#结果-2)
|
||||
- [用法](#用法)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
|
|
@ -365,8 +365,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATA_PATH] [ANNO_PATH] [DEVICE_ID]
|
|||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| 模型版本 | YOLOv3_Resnet18 V1 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-01 |
|
||||
| MindSpore版本 | 0.2.0-alpha |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | COCO2017 |
|
||||
| 训练参数 | epoch = 150, batch_size = 32, lr = 0.001 |
|
||||
| 优化器 | Adam |
|
||||
|
|
@ -383,8 +383,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATA_PATH] [ANNO_PATH] [DEVICE_ID]
|
|||
| ------------------- | ----------------------------------------------- |
|
||||
| 模型版本 | YOLOv3_Resnet18 V1 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-01 |
|
||||
| MindSpore版本 | 0.2.0-alpha |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | COCO2017 |
|
||||
| batch_size | 1 |
|
||||
| 输出 | 精度和召回 |
|
||||
|
|
|
|||
|
|
@ -1,5 +1,6 @@
|
|||
# 目录
|
||||
|
||||
- [目录](#目录)
|
||||
- [YOLOv4说明](#yolov4说明)
|
||||
- [模型架构](#模型架构)
|
||||
- [预训练模型](#预训练模型)
|
||||
|
|
@ -7,18 +8,25 @@
|
|||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本和示例代码](#脚本和示例代码)
|
||||
- [脚本和示例代码](#脚本和示例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练)
|
||||
- [分布式训练](#分布式训练)
|
||||
- [迁移学习](#迁移学习)
|
||||
- [评估过程](#评估过程)
|
||||
- [评估](#评估)
|
||||
- [验证](#验证)
|
||||
- [Test-dev](#test-dev)
|
||||
- [转换过程](#转换过程)
|
||||
- [转换](#转换)
|
||||
- [推理过程](#推理过程)
|
||||
- [用法](#用法)
|
||||
- [结果](#结果)
|
||||
- [模型说明](#模型说明)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [推理性能](#推理性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
# [YOLOv4说明](#目录)
|
||||
|
|
|
|||
|
|
@ -2,28 +2,27 @@
|
|||
|
||||
<!--TOC -->
|
||||
|
||||
- [Graph Attention Networks Description](#graph-attention-networks-description)
|
||||
- [Model architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Contents](#contents)
|
||||
- [Graph Attention Networks Description](#graph-attention-networks-description)
|
||||
- [Model architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Training](#training)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#evaluation-performance)
|
||||
- [Description of random situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Description of random situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
<!--TOC -->
|
||||
|
||||
|
|
@ -273,8 +272,8 @@ test acc=0.84199995
|
|||
| Parameter | GAT |
|
||||
| ------------------------------------ | ----------------------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| uploaded Date | 06/16/2020(month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021(month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | Cora/Citeseer |
|
||||
| Training Parameter | epoch=200 |
|
||||
| Optimizer | Adam |
|
||||
|
|
|
|||
|
|
@ -3,26 +3,26 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [图注意力网络描述](#图注意力网络描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [特性](#特性)
|
||||
- [混合精度](#混合精度)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练)
|
||||
- [图注意力网络描述](#图注意力网络描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [特性](#特性)
|
||||
- [混合精度](#混合精度)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [训练](#训练)
|
||||
- [推理过程](#推理过程)
|
||||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [result](#result)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -269,8 +269,8 @@ test acc=0.84199995
|
|||
| 参数 | GAT |
|
||||
| ------------------------------------ | ----------------------------------------- |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-16 |
|
||||
| MindSpore版本 | 0.5.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | Cora/Citeseer |
|
||||
| 训练参数 | epoch=200 |
|
||||
| 优化器 | Adam |
|
||||
|
|
|
|||
|
|
@ -1,22 +1,28 @@
|
|||
# Contents
|
||||
|
||||
- [GCN Description](#gcn-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Contents](#contents)
|
||||
- [GCN Description](#gcn-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Usage](#usage)
|
||||
- [Launch](#launch)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training, Evaluation, Test Process](#training-evaluation-test-process)
|
||||
- [Training, Evaluation, Test Process](#training-evaluation-test-process)
|
||||
- [Usage](#usage-1)
|
||||
- [Launch](#launch-1)
|
||||
- [Result](#result)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [result](#result-1)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## [GCN Description](#contents)
|
||||
|
||||
|
|
@ -253,8 +259,8 @@ Test set results: accuracy= 0.81300
|
|||
| Parameters | GCN |
|
||||
| -------------------------- | -------------------------------------------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| uploaded Date | 06/09/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | Cora/Citeseer |
|
||||
| Training Parameters | epoch=200 |
|
||||
| Optimizer | Adam |
|
||||
|
|
|
|||
|
|
@ -3,28 +3,28 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [图卷积网络描述](#图卷积网络描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [用法](#用法)
|
||||
- [启动](#启动)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [培训、评估、测试过程](#培训评估测试过程)
|
||||
- [用法](#用法-1)
|
||||
- [启动](#启动-1)
|
||||
- [结果](#结果)
|
||||
- [图卷积网络描述](#图卷积网络描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [用法](#用法)
|
||||
- [启动](#启动)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [培训、评估、测试过程](#培训评估测试过程)
|
||||
- [用法](#用法-1)
|
||||
- [启动](#启动-1)
|
||||
- [结果](#结果)
|
||||
- [推理过程](#推理过程)
|
||||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [result](#result)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -261,8 +261,8 @@ Test set results: accuracy= 0.81300
|
|||
| 参数 | GCN |
|
||||
| -------------------------- | -------------------------------------------------------------- |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-09 |
|
||||
| MindSpore版本 | 0.5.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | Cora/Citeseer |
|
||||
| 训练参数 | epoch=200 |
|
||||
| 优化器 | Adam |
|
||||
|
|
|
|||
|
|
@ -29,8 +29,10 @@
|
|||
- [evaluation on cluener dataset when running on Ascend](#evaluation-on-cluener-dataset-when-running-on-ascend)
|
||||
- [evaluation on msra dataset when running on Ascend](#evaluation-on-msra-dataset-when-running-on-ascend)
|
||||
- [evaluation on squad v1.1 dataset when running on Ascend](#evaluation-on-squad-v11-dataset-when-running-on-ascend)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Usage](#usage)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Pretraining Performance](#pretraining-performance)
|
||||
|
|
@ -720,18 +722,18 @@ F1 0.931243
|
|||
| -------------------------- | ---------------------------------------------------------- | ------------------------- |
|
||||
| Model Version | BERT_base | BERT_base |
|
||||
| Resource | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 | NV SMX2 V100-16G, cpu: Intel(R) Xeon(R) Platinum 8160 CPU @2.10GHz, memory: 256G |
|
||||
| uploaded Date | 08/22/2020 | 05/06/2020 |
|
||||
| MindSpore Version | 1.0.0 | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | cn-wiki-128(4000w) | cn-wiki-128(4000w) |
|
||||
| Training Parameters | src/config.py | src/config.py |
|
||||
| Training Parameters | pretrain_config.yaml | pretrain_config.yaml |
|
||||
| Optimizer | Lamb | AdamWeightDecay |
|
||||
| Loss Function | SoftmaxCrossEntropy | SoftmaxCrossEntropy |
|
||||
| outputs | probability | probability |
|
||||
| Epoch | 40 | 40 |
|
||||
| Batch_size | 256*8 | 32*8 |
|
||||
| Loss | 1.7 | 1.7 |
|
||||
| Speed | 340ms/step | 290ms/step |
|
||||
| Total time | 73h | 610H |
|
||||
| Speed | 284ms/step | 180ms/step |
|
||||
| Total time | 63H | 610H |
|
||||
| Params (M) | 110M | 110M |
|
||||
| Checkpoint for Fine tuning | 1.2G(.ckpt file) | 1.2G(.ckpt file) |
|
||||
| Scripts | [BERT_base](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/nlp/bert) | [BERT_base](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/nlp/bert) |
|
||||
|
|
@ -740,8 +742,8 @@ F1 0.931243
|
|||
| -------------------------- | ---------------------------------------------------------- |
|
||||
| Model Version | BERT_NEZHA |
|
||||
| Resource | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 08/20/2020 |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | cn-wiki-128(4000w) |
|
||||
| Training Parameters | src/config.py |
|
||||
| Optimizer | Lamb |
|
||||
|
|
@ -750,8 +752,8 @@ F1 0.931243
|
|||
| Epoch | 40 |
|
||||
| Batch_size | 96*8 |
|
||||
| Loss | 1.7 |
|
||||
| Speed | 360ms/step |
|
||||
| Total time | 200h |
|
||||
| Speed | 320ms/step |
|
||||
| Total time | 180h |
|
||||
| Params (M) | 340M |
|
||||
| Checkpoint for Fine tuning | 3.2G(.ckpt file) |
|
||||
| Scripts | [BERT_NEZHA](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/nlp/bert) |
|
||||
|
|
@ -762,8 +764,8 @@ F1 0.931243
|
|||
| -------------------------- | ----------------------------- |
|
||||
| Model Version | |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| uploaded Date | 08/22/2020 |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | cola, 1.2W |
|
||||
| batch_size | 32(1P) |
|
||||
| Accuracy | 0.588986 |
|
||||
|
|
|
|||
|
|
@ -33,7 +33,7 @@
|
|||
- [导出mindir模型](#导出mindir模型)
|
||||
- [推理过程](#推理过程)
|
||||
- [用法](#用法-2)
|
||||
- [结果](#结果-2)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [预训练性能](#预训练性能)
|
||||
|
|
@ -682,54 +682,54 @@ F1 0.931243
|
|||
| -------------------------- | ---------------------------------------------------------- | ------------------------- |
|
||||
| 模型版本 | BERT_base | BERT_base |
|
||||
| 资源 |Ascend 910；CPU 2.60GHz，192核；内存 755GB；系统 Euler2.8 || NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-08-22 | 2020-05-06 |
|
||||
| MindSpore版本 | 0.6.0 | 0.3.0 |
|
||||
| 数据集 | cn-wiki-128(4000w) | ImageNet |
|
||||
| 训练参数 | src/gd_config.py | src/gd_config.py |
|
||||
| 优化器 | Lamb | Momentum |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | cn-wiki-128(4000w) | cn-wiki-128 |
|
||||
| 训练参数 | pretrain_config.yaml | pretrain_config.yaml |
|
||||
| 优化器 | Lamb | Lamb |
|
||||
| 损失函数 | SoftmaxCrossEntropy | SoftmaxCrossEntropy |
|
||||
| 输出 | 概率 | |
|
||||
| 轮次 | 40 | | |
|
||||
| Batch_size | 256*8 | 130（8卡） | |
|
||||
| Batch_size | 256*8 | 32*8 |
|
||||
| 损失 | 1.7 | 1.913 |
|
||||
| 速度 | 340毫秒/步 | 1.913 |
|
||||
| 总时长 | 73小时 | |
|
||||
| 速度 | 284毫秒/步 |180毫秒/步 |
|
||||
| 总时长 | 63小时 | |
|
||||
| 参数（M） | 110 | |
|
||||
| 微调检查点 | 1.2G（.ckpt文件） | |
|
||||
|
||||
| 参数 | Ascend | GPU |
|
||||
| -------------------------- | ---------------------------------------------------------- | ------------------------- |
|
||||
| 模型版本 | BERT_NEZHA | BERT_NEZHA |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755GB；系统 Euler2.8 || NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-08-20 | 2020-05-06 |
|
||||
| MindSpore版本 | 0.6.0 | 0.3.0 |
|
||||
| 数据集 | cn-wiki-128(4000w) | ImageNet |
|
||||
| 训练参数 | src/config.py | src/config.py |
|
||||
| 优化器 | Lamb | Momentum |
|
||||
| 损失函数 | SoftmaxCrossEntropy | SoftmaxCrossEntropy |
|
||||
| 输出 | 概率 | |
|
||||
| 轮次 | 40 | | |
|
||||
| Batch_size | 96*8 | 130（8卡） |
|
||||
| 损失 | 1.7 | 1.913 |
|
||||
| 速度 | 360毫秒/步 | 1.913 |
|
||||
| 总时长 | 200小时 |
|
||||
| 参数（M） | 340 | |
|
||||
| 微调检查点 | 3.2G（.ckpt文件） | |
|
||||
| 参数 | Ascend |
|
||||
| -------------------------- | ---------------------------------------------------------- |
|
||||
| 模型版本 | BERT_NEZHA |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755GB；系统 Euler2.8 |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | cn-wiki-128(4000w) |
|
||||
| 训练参数 | pretrain_config.yaml |
|
||||
| 优化器 | Lamb |
|
||||
| 损失函数 | SoftmaxCrossEntropy |
|
||||
| 输出 | 概率 |
|
||||
| 轮次 | 40 |
|
||||
| Batch_size | 96*8 |
|
||||
| 损失 | 1.7 |
|
||||
| 速度 | 320毫秒/步 |
|
||||
| 总时长 | 180小时 |
|
||||
| 参数（M） | 340 |
|
||||
| 微调检查点 | 3.2G（.ckpt文件） |
|
||||
|
||||
#### 推理性能
|
||||
|
||||
| 参数 | Ascend | GPU |
|
||||
| -------------------------- | ----------------------------- | ------------------------- |
|
||||
| 模型版本 | | |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-08-22 | 2020-05-22 |
|
||||
| MindSpore版本 | 0.6.0 | 0.2.0 |
|
||||
| 数据集 | cola，1.2W | ImageNet, 1.2W |
|
||||
| batch_size | 32（单卡） | 130（8卡） |
|
||||
| 准确率 | 0.588986 | ACC1[72.07%] ACC5[90.90%] |
|
||||
| 速度 | 59.25毫秒/步 | |
|
||||
| 总时长 | 15分钟 | |
|
||||
| 推理模型 | 1.2G（.ckpt文件） | |
|
||||
| 参数 | Ascend |
|
||||
| -------------------------- | ----------------------------- |
|
||||
| 模型版本 | |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | cola，1.2W |
|
||||
| batch_size | 32（单卡） |
|
||||
| 准确率 | 0.588986 |
|
||||
| 速度 | 59.25毫秒/步 |
|
||||
| 总时长 | 15分钟 |
|
||||
| 推理模型 | 1.2G（.ckpt文件） |
|
||||
|
||||
# 随机情况说明
|
||||
|
||||
|
|
|
|||
|
|
@ -1,20 +1,23 @@
|
|||
# Bert-THOR Example
|
||||
|
||||
- [Description](#Description)
|
||||
- [Model Architecture](#Model-Architecture)
|
||||
- [Dataset](#Dataset)
|
||||
- [Features](#Features)
|
||||
- [Environment Requirements](#Environment-Requirements)
|
||||
- [Quick Start](#Quick-Start)
|
||||
- [Script Description](#Script-Description)
|
||||
- [Script and Sample Code](#Script-Code-Structure)
|
||||
- [Script Parameters](#Script-Parameters)
|
||||
- [Training Process](#Training-Process)
|
||||
- [Evaluation Process](#Evaluation-Process)
|
||||
- [Model Description](#Model-Description)
|
||||
- [Evaluation Performance](#Evaluation-Performance)
|
||||
- [Description of Random Situation](#Description-of-Random-Situation)
|
||||
- [ModelZoo Homepage](#ModelZoo-Homepage)
|
||||
- [Bert-THOR Example](#bert-thor-example)
|
||||
- [Description](#description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script Code Structure](#script-code-structure)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Ascend 910](#ascend-910)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Ascend910](#ascend910)
|
||||
- [Model Description](#model-description)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## Description
|
||||
|
||||
|
|
@ -205,16 +208,16 @@ step: 3000 Accuracy: [0.71377236]
|
|||
| -------------------------- | -------------------------------------- |
|
||||
| Model Version | BERT-LARGE |
|
||||
| Resource | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 08/20/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.6.0-alpha |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | MLPerf v0.7 dataset |
|
||||
| Training Parameters | total steps=3000, batch_size = 12 |
|
||||
| Training Parameters | total steps=3000, batch_size = 12*8 |
|
||||
| Optimizer | THOR |
|
||||
| Loss Function | Softmax Cross Entropy |
|
||||
| outputs | probability |
|
||||
| Loss |1.5654222 |
|
||||
| Speed | 275ms/step（8pcs） |
|
||||
| Total time | 14 mins |
|
||||
| Speed | 218ms/step（8pcs） |
|
||||
| Total time | 11 mins |
|
||||
| Parameters (M) | 330 |
|
||||
| Checkpoint for Fine tuning | 4.5G(.ckpt file) |
|
||||
| Scripts | https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/nlp/bert_thor |
|
||||
|
|
|
|||
|
|
@ -14,8 +14,8 @@
|
|||
- [训练过程](#训练过程)
|
||||
- [Ascend 910](#ascend-910)
|
||||
- [评估过程](#评估过程)
|
||||
- [Ascend 910](#ascend-910-1)
|
||||
- [模型描述](#模型描述)
|
||||
- [Ascend910](#ascend910)
|
||||
- [模型描述](#模型描述)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo首页](#modelzoo首页)
|
||||
|
|
@ -212,16 +212,16 @@ step: 3000 Accuracy: [0.71377236]
|
|||
| -------------------------- | -------------------------------------- |
|
||||
| 模型版本 | BERT-LARGE |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755GB；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-08-20 |
|
||||
| MindSpore版本 | 0.6.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | MLPerf v0.7 |
|
||||
| 训练参数 |总步数=3000，batch_size=12 |
|
||||
| 训练参数 |总步数=3000，batch_size=12*8 |
|
||||
| 优化器 | THOR |
|
||||
| 损失函数 | Softmax Cross Entropy |
|
||||
| 输出 | 概率 |
|
||||
| 损失 | 1.5654222 |
|
||||
| 速度 | 275毫秒/步|
|
||||
| 总时长 | 14分钟 |
|
||||
| 速度 | 218毫秒/步|
|
||||
| 总时长 | 11分钟 |
|
||||
| 参数（M） | 330 |
|
||||
| 微调检查点 | 4.5G （.ckpt文件） |
|
||||
| 脚本 | https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/nlp/bert_thor |
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
[查看中文](./README_CN.md)
|
||||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [LSTM Description](#lstm-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
|
|
@ -9,11 +10,16 @@
|
|||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Script Parameters](#training-script-parameters)
|
||||
- [Running Options](#running-options)
|
||||
- [Network Parameters](#network-parameters)
|
||||
- [Dataset Preparation](#dataset-preparation)
|
||||
- [Training Process](#training-process)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Usage](#usage)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
|
|
@ -427,8 +433,8 @@ Inference result is saved in current path, you can find result in acc.log file.
|
|||
| Parameters | LSTM (Ascend) | LSTM (GPU) | LSTM (CPU) |
|
||||
| -------------------------- | -------------------------- | -------------------------------------------------------------- | -------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 | Tesla V100-SMX2-16GB | Ubuntu X86-i7-8565U-16GB |
|
||||
| uploaded Date | 12/21/2020 (month/day/year)| 10/28/2020 (month/day/year) | 10/28/2020 (month/day/year)|
|
||||
| MindSpore Version | 1.1.0 | 1.0.0 | 1.0.0 |
|
||||
| uploaded Date | 12/21/2020 (month/day/year)| 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year)|
|
||||
| MindSpore Version | 1.1.0 | 1.3.0 | 1.3.0 |
|
||||
| Dataset | aclimdb_v1 | aclimdb_v1 | aclimdb_v1 |
|
||||
| Training Parameters | epoch=20, batch_size=64 | epoch=20, batch_size=64 | epoch=20, batch_size=64 |
|
||||
| Optimizer | Momentum | Momentum | Momentum |
|
||||
|
|
@ -444,8 +450,8 @@ Inference result is saved in current path, you can find result in acc.log file.
|
|||
| Parameters | LSTM (Ascend) | LSTM (GPU) | LSTM (CPU) |
|
||||
| ------------------- | ---------------------------- | --------------------------- | ---------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 | Tesla V100-SMX2-16GB | Ubuntu X86-i7-8565U-16GB |
|
||||
| uploaded Date | 12/21/2020 (month/day/year) | 10/28/2020 (month/day/year) | 10/28/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.1.0 | 1.0.0 | 1.0.0 |
|
||||
| uploaded Date | 12/21/2020 (month/day/year) | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.1.0 | 1.3.0 | 1.3.0 |
|
||||
| Dataset | aclimdb_v1 | aclimdb_v1 | aclimdb_v1 |
|
||||
| batch_size | 64 | 64 | 64 |
|
||||
| Accuracy | 85% | 84% | 83% |
|
||||
|
|
|
|||
|
|
@ -19,8 +19,8 @@
|
|||
- [评估过程](#评估过程)
|
||||
- [导出mindir模型](#导出mindir模型)
|
||||
- [推理过程](#推理过程)
|
||||
- [用法](#用法-2)
|
||||
- [结果](#结果-2)
|
||||
- [用法](#用法)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
|
|
@ -433,8 +433,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATASET_PATH] [NEED_PREPROCESS] [DEVICE_TAR
|
|||
| 参数 | LSTM (Ascend) | LSTM (GPU) | LSTM (CPU) |
|
||||
| -------------------------- | -------------------------- | -------------------------------------------------------------- | -------------------------- |
|
||||
| 资源 | Ascend 910 | Tesla V100-SMX2-16GB | Ubuntu X86-i7-8565U-16GB |
|
||||
| 上传日期 | 2020-12-21 | 2020-08-06 | 2020-08-06 |
|
||||
| MindSpore版本 | 1.1.0 | 0.6.0-beta | 0.6.0-beta |
|
||||
| 上传日期 | 2020-12-21 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.1.0 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | aclimdb_v1 | aclimdb_v1 | aclimdb_v1 |
|
||||
| 训练参数 | epoch=20, batch_size=64 | epoch=20, batch_size=64 | epoch=20, batch_size=64 |
|
||||
| 优化器 | Momentum | Momentum | Momentum |
|
||||
|
|
@ -450,8 +450,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATASET_PATH] [NEED_PREPROCESS] [DEVICE_TAR
|
|||
| 参数 | LSTM (Ascend) | LSTM (GPU) | LSTM (CPU) |
|
||||
| ------------------- | ---------------------------- | --------------------------- | ---------------------------- |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | Tesla V100-SMX2-16GB | Ubuntu X86-i7-8565U-16GB |
|
||||
| 上传日期 | 2020-12-21 | 2020-08-06 | 2020-08-06 |
|
||||
| MindSpore版本 | 1.1.0 | 0.6.0-beta | 0.6.0-beta |
|
||||
| 上传日期 | 2020-12-21 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.1.0 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | aclimdb_v1 | aclimdb_v1 | aclimdb_v1 |
|
||||
| batch_size | 64 | 64 | 64 |
|
||||
| 准确率 | 85% | 84% | 83% |
|
||||
|
|
|
|||
|
|
@ -2,6 +2,7 @@
|
|||
|
||||
<!-- TOC -->
|
||||
|
||||
- [Contexts](#contexts)
|
||||
- [MASS: Masked Sequence to Sequence Pre-training for Language Generation Description](#mass-masked-sequence-to-sequence-pre-training-for-language-generation-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
|
|
@ -694,8 +695,8 @@ The comparisons between MASS and other baseline methods in terms of PPL on Corne
|
|||
|:---------------------------|:--------------------------------------------------------------------------|
|
||||
| Model Version | v1 |
|
||||
| Resource | Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 |
|
||||
| uploaded Date | 05/24/2020 |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 06/21/2021 |
|
||||
| MindSpore Version | 1.2.1 |
|
||||
| Dataset | News Crawl 2007-2017 English monolingual corpus, Gigaword corpus, Cornell Movie Dialog corpus |
|
||||
| Training Parameters | Epoch=50, steps=XXX, batch_size=192, lr=1e-4 |
|
||||
| Optimizer | Adam |
|
||||
|
|
@ -713,8 +714,8 @@ The comparisons between MASS and other baseline methods in terms of PPL on Corne
|
|||
|:---------------------------|:-----------------------------------------------------------|
|
||||
| Model Version | V1 |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| uploaded Date | 05/24/2020 |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 06/21/2021 |
|
||||
| MindSpore Version | 1.2.1 |
|
||||
| Dataset | Gigaword corpus, Cornell Movie Dialog corpus |
|
||||
| batch_size | --- |
|
||||
| outputs | Sentence and probability |
|
||||
|
|
|
|||
|
|
@ -27,12 +27,12 @@
|
|||
- [预训练](#预训练)
|
||||
- [微调](#微调)
|
||||
- [推理](#推理)
|
||||
- [Mindir推理](#Mindir推理)
|
||||
- [Mindir推理](#mindir推理)
|
||||
- [导出模型](#导出模型)
|
||||
- [在Ascend310执行推理](#在Ascend310执行推理)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [性能](#性能)
|
||||
- [结果](#结果)
|
||||
- [结果](#结果-1)
|
||||
- [文本摘要微调](#文本摘要微调)
|
||||
- [会话应答微调](#会话应答微调)
|
||||
- [训练性能](#训练性能)
|
||||
|
|
@ -696,8 +696,8 @@ bash run_infer_310.sh [MINDIR_PATH] [CONFIG] [VOCAB] [OUTPUT] [NEED_PREPROCESS]
|
|||
|:---------------------------|:--------------------------------------------------------------------------|
|
||||
| 模型版本 | v1 |
|
||||
| 资源 | Ascend 910；CPU 2.60GHz，192核；内存 755GB；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-05-24 |
|
||||
| MindSpore版本 | 0.2.0 |
|
||||
| 上传日期 | 2021-06-21 |
|
||||
| MindSpore版本 | 1.2.1 |
|
||||
| 数据集 | News Crawl 2007-2017英语单语语料库、Gigaword语料库、Cornell电影对白语料库 |
|
||||
| 训练参数 | Epoch=50, steps=XXX, batch_size=192, lr=1e-4 |
|
||||
| 优化器 | Adam |
|
||||
|
|
@ -715,8 +715,8 @@ bash run_infer_310.sh [MINDIR_PATH] [CONFIG] [VOCAB] [OUTPUT] [NEED_PREPROCESS]
|
|||
|:---------------------------|:-----------------------------------------------------------|
|
||||
|模型版本| V1 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-05-24 |
|
||||
| MindSpore版本 | 0.2.0 |
|
||||
| 上传日期 | 2020-06-21 |
|
||||
| MindSpore版本 | 1.2.1 |
|
||||
| 数据集 | Gigaword语料库、Cornell电影对白语料库 |
|
||||
| batch_size | --- |
|
||||
| 输出 | 句子及概率 |
|
||||
|
|
|
|||
|
|
@ -12,8 +12,8 @@
|
|||
- [General Distill](#general-distill)
|
||||
- [Task Distill](#task-distill)
|
||||
- [Options and Parameters](#options-and-parameters)
|
||||
- [Options:](#options)
|
||||
- [Parameters:](#parameters)
|
||||
- [Options](#options)
|
||||
- [Parameters](#parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [running on Ascend](#running-on-ascend)
|
||||
|
|
@ -557,8 +557,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| -------------------------- | ---------------------------------------------------------- | ------------------------- |
|
||||
| Model Version | TinyBERT | TinyBERT |
|
||||
| Resource |Ascend 910; cpu 2.60GHz, 192cores; memory 755G; OS Euler2.8 | NV SMX2 V100-32G, cpu:2.10GHz 64cores, memory:251G |
|
||||
| uploaded Date | 08/20/2020 | 08/24/2020 |
|
||||
| MindSpore Version | 1.0.0 | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | en-wiki-128 | en-wiki-128 |
|
||||
| Training Parameters | src/gd_config.yaml | src/gd_config.yaml |
|
||||
| Optimizer | AdamWeightDecay | AdamWeightDecay |
|
||||
|
|
@ -577,8 +577,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| -------------------------- | ----------------------------- | ------------------------- |
|
||||
| Model Version | | |
|
||||
| Resource | Ascend 910; OS Euler2.8 | NV SMX2 V100-32G |
|
||||
| uploaded Date | 08/20/2020 | 08/24/2020 |
|
||||
| MindSpore Version | 1.0.0 | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 | 07/05/2021 |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | SST-2, | SST-2 |
|
||||
| batch_size | 32 | 32 |
|
||||
| Accuracy | 0.902777 | 0.9086 |
|
||||
|
|
|
|||
|
|
@ -556,8 +556,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATASET_PATH] [SCHEMA_DIR] [DATASET_TYPE] [
|
|||
| -------------------------- | ---------------------------------------------------------- | ------------------------- |
|
||||
| 模型版本 | TinyBERT | TinyBERT |
|
||||
| 资源 | Ascend 910；cpu 2.60GHz，192核；内存 755G；系统 Euler2.8 | NV SMX2 V100-32G, cpu:2.10GHz 64核, 内存:251G |
|
||||
| 上传日期 | 2020-08-20 | 2020-08-24 |
|
||||
| MindSpore版本 | 0.6.0 | 0.7.0 |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | en-wiki-128 | en-wiki-128 |
|
||||
| 训练参数 | src/gd_config.yaml | src/gd_config.yaml |
|
||||
| 优化器| AdamWeightDecay | AdamWeightDecay |
|
||||
|
|
@ -575,8 +575,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATASET_PATH] [SCHEMA_DIR] [DATASET_TYPE] [
|
|||
| -------------------------- | ----------------------------- | ------------------------- |
|
||||
| 模型版本 | | |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | NV SMX2 V100-32G |
|
||||
| 上传日期 | 2020-08-20 | 2020-08-24 |
|
||||
| MindSpore版本 | 0.6.0 | 0.7.0 |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | SST-2, | SST-2 |
|
||||
| batch_size | 32 | 32 |
|
||||
| 准确率 | 0.902777 | 0.9086 |
|
||||
|
|
|
|||
|
|
@ -2,14 +2,18 @@
|
|||
|
||||
[查看中文](./README_CN.md)
|
||||
|
||||
- [Transformer Description](#transformer-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Contents](#contents)
|
||||
- [Transformer Description](#transformer-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Script Parameters](#training-script-parameters)
|
||||
- [Running Options](#running-options)
|
||||
- [Network Parameters](#network-parameters)
|
||||
- [Dataset Preparation](#dataset-preparation)
|
||||
- [Training Process](#training-process)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
|
|
@ -17,12 +21,12 @@
|
|||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## [Transformer Description](#contents)
|
||||
|
||||
|
|
@ -385,8 +389,8 @@ Inference result is saved in current path, 'output_file' will generate in path s
|
|||
| Parameters | Ascend |
|
||||
| -------------------------- | -------------------------------------------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| uploaded Date | 09/15/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | WMT Englis-German |
|
||||
| Training Parameters | epoch=52, batch_size=96 |
|
||||
| Optimizer | Adam |
|
||||
|
|
@ -403,8 +407,8 @@ Inference result is saved in current path, 'output_file' will generate in path s
|
|||
| Parameters | Ascend |
|
||||
| ------------------- | --------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 09/15/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | WMT newstest2014 |
|
||||
| batch_size | 1 |
|
||||
| outputs | BLEU score |
|
||||
|
|
|
|||
|
|
@ -5,30 +5,30 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [Transformer 概述](#transfomer-概述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本和样例代码](#脚本和样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练脚本参数](#训练脚本参数)
|
||||
- [运行选项](#运行选项)
|
||||
- [网络参数](#网络参数)
|
||||
- [准备数据集](#准备数据集)
|
||||
- [训练过程](#训练过程)
|
||||
- [评估过程](#评估过程)
|
||||
- [Transformer 概述](#transformer-概述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本和样例代码](#脚本和样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练脚本参数](#训练脚本参数)
|
||||
- [运行选项](#运行选项)
|
||||
- [网络参数](#网络参数)
|
||||
- [准备数据集](#准备数据集)
|
||||
- [训练过程](#训练过程)
|
||||
- [评估过程](#评估过程)
|
||||
- [推理过程](#推理过程)
|
||||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -392,8 +392,8 @@ bash run_infer_310.sh [MINDIR_PATH] [NEED_PREPROCESS] [DEVICE_ID]
|
|||
| 参数 | Ascend |
|
||||
| -------------------------- | -------------------------------------------------------------- |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-09 |
|
||||
| MindSpore版本 | 0.5.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | WMT英-德翻译数据集 |
|
||||
| 训练参数 | epoch=52, batch_size=96 |
|
||||
| 优化器 | Adam |
|
||||
|
|
@ -410,8 +410,8 @@ bash run_infer_310.sh [MINDIR_PATH] [NEED_PREPROCESS] [DEVICE_ID]
|
|||
| 参数 | Ascend |
|
||||
| ------------------- | --------------------------- |
|
||||
|资源| Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-06-09 |
|
||||
| MindSpore版本 | 0.5.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | WMT newstest2014 |
|
||||
| batch_size | 1 |
|
||||
| 输出 | BLEU score |
|
||||
|
|
|
|||
|
|
@ -1,5 +1,6 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [DeepFM Description](#deepfm-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
|
|
@ -10,7 +11,7 @@
|
|||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Distributed Training](#distributed-training)
|
||||
- [Distributed Training](#distributed-training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Inference Process](#inference-process)
|
||||
|
|
@ -19,8 +20,8 @@
|
|||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#evaluation-performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -417,15 +418,15 @@ auc : 0.8057789065281104
|
|||
| -------------------------- | ----------------------------------------------------------- | ---------------------- |
|
||||
| Model Version | DeepFM | To do |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 | To do |
|
||||
| uploaded Date | 09/15/2020 (month/day/year) | To do |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) | To do |
|
||||
| MindSpore Version | 1.0.0 | To do |
|
||||
| Dataset | [1] | To do |
|
||||
| Training Parameters | epoch=15, batch_size=1000, lr=1e-5 | To do |
|
||||
| Training Parameters | epoch=15, batch_size=16000, lr=1e-5 | To do |
|
||||
| Optimizer | Adam | To do |
|
||||
| Loss Function | Sigmoid Cross Entropy With Logits | To do |
|
||||
| outputs | Accuracy | To do |
|
||||
| Loss | 0.45 | To do |
|
||||
| Speed | 1pc: 8.16 ms/step; | To do |
|
||||
| Speed | 1pc: 21 ms/step; | To do |
|
||||
| Total time | 1pc: 90 mins; | To do |
|
||||
| Parameters (M) | 16.5 | To do |
|
||||
| Checkpoint for Fine tuning | 190M (.ckpt file) | To do |
|
||||
|
|
@ -437,8 +438,8 @@ auc : 0.8057789065281104
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| Model Version | DeepFM | To do |
|
||||
| Resource | Ascend 910; OS Euler2.8 | To do |
|
||||
| Uploaded Date | 05/27/2020 (month/day/year) | To do |
|
||||
| MindSpore Version | 0.3.0-alpha | To do |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) | To do |
|
||||
| MindSpore Version | 1.3.0 | To do |
|
||||
| Dataset | [1] | To do |
|
||||
| batch_size | 1000 | To do |
|
||||
| outputs | accuracy | To do |
|
||||
|
|
|
|||
|
|
@ -3,11 +3,11 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [DeepFM概述](#deepfm概述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [DeepFM概述](#deepfm概述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本和样例代码](#脚本和样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
|
|
@ -20,7 +20,7 @@
|
|||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [推理性能](#推理性能)
|
||||
|
|
@ -399,15 +399,15 @@ auc : 0.8057789065281104
|
|||
| -------------------------- | ----------------------------------------------------------- | ---------------------- |
|
||||
| 模型版本 | DeepFM | 待运行 |
|
||||
| 资源 |Ascend 910；CPU 2.60GHz，192核；内存 755G；系统 Euler2.8 | 待运行 |
|
||||
| 上传日期 | 2020-05-17 | 待运行 |
|
||||
| MindSpore版本 | 0.3.0-alpha | 待运行 |
|
||||
| 上传日期 | 2021-07-05 | 待运行 |
|
||||
| MindSpore版本 | 1.3.0 | 待运行 |
|
||||
| 数据集 | [1] | 待运行 |
|
||||
| 训练参数 | epoch=15, batch_size=1000, lr=1e-5 | 待运行 |
|
||||
| 训练参数 | epoch=15, batch_size=16000, lr=1e-5 | 待运行 |
|
||||
| 优化器 | Adam | 待运行 |
|
||||
| 损失函数 | Sigmoid Cross Entropy With Logits | 待运行 |
|
||||
| 输出 | 准确率 | 待运行 |
|
||||
| 损失 | 0.45 | 待运行 |
|
||||
| 速度| 单卡：8.16毫秒/步; | 待运行 |
|
||||
| 速度| 单卡：21毫秒/步; | 待运行 |
|
||||
| 总时长| 单卡：90 分钟; | 待运行 |
|
||||
| 参数(M) | 16.5 | 待运行 |
|
||||
| 微调检查点 | 190M (.ckpt 文件) | 待运行 |
|
||||
|
|
@ -419,8 +419,8 @@ auc : 0.8057789065281104
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| 模型版本 | DeepFM | 待运行 |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 | 待运行 |
|
||||
| 上传日期 | 2020-05-27 | 待运行 |
|
||||
| MindSpore版本 | 0.3.0-alpha | 待运行 |
|
||||
| 上传日期 | 2021-07-05 | 待运行 |
|
||||
| MindSpore版本 | 1.3.0 | 待运行 |
|
||||
| 数据集 | [1] | 待运行 |
|
||||
| batch_size | 1000 | 待运行 |
|
||||
| 输出 | 准确率 | 待运行 |
|
||||
|
|
|
|||
|
|
@ -9,15 +9,16 @@
|
|||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Script Parameters](#training-script-parameters)
|
||||
- [Preprocess Script Parameters](#preprocess-script-parameters)
|
||||
- [Training Script Parameters](#training-script-parameters)
|
||||
- [Preprocess Script Parameters](#preprocess-script-parameters)
|
||||
- [Dataset Preparation](#dataset-preparation)
|
||||
- [Process the Real World Data](#process-the-real-world-data)
|
||||
- [Generate and Process the Synthetic Data](#generate-and-process-the-synthetic-data)
|
||||
- [Process the Real World Data](#process-the-real-world-data)
|
||||
- [Generate and Process the Synthetic Data](#generate-and-process-the-synthetic-data)
|
||||
- [Training Process](#training-process)
|
||||
- [SingleDevice](#singledevice)
|
||||
- [Distribute Training](#distribute-training)
|
||||
- [Parameter Server](#parameter-server)
|
||||
- [SingleDevice](#singledevice)
|
||||
- [SingleDevice For Cache Mode](#singledevice-for-cache-mode)
|
||||
- [Distribute Training](#distribute-training)
|
||||
- [Parameter Server](#parameter-server)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
|
|
@ -25,8 +26,9 @@
|
|||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Ultimate performance experience](#ultimate-performance-experience)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -444,8 +446,8 @@ Inference result is saved in current path, you can find result like this in acc.
|
|||
| Parameters | Single <br />Ascend | Single<br />GPU | Data-Parallel-8P | Host-Device-mode-8P |
|
||||
| ------------------------ | ------------------------------- | ------------------------------- | ------------------------------- | ------------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 | Tesla V100-PCIE 32G | Ascend 910; OS Euler2.8 | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 08/21/2020 (month/day/year) | 08/21/2020 (month/day/year) | 08/21/2020 (month/day/year) | 08/21/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0 | 1.0 | 1.0 | 1.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 | 1.3.0 | 1.3.0 |
|
||||
| Dataset | [1] | [1] | [1] | [1] |
|
||||
| Training Parameters | Epoch=15,<br />batch_size=16000 | Epoch=15,<br />batch_size=16000 | Epoch=15,<br />batch_size=16000 | Epoch=15,<br />batch_size=16000 |
|
||||
| Optimizer | FTRL,Adam | FTRL,Adam | FTRL,Adam | FTRL,Adam |
|
||||
|
|
@ -465,8 +467,8 @@ Note: The result of GPU is tested under the master version. The parameter server
|
|||
| Parameters | Wide&Deep |
|
||||
| ----------------- | --------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 10/27/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | [1] |
|
||||
| Batch Size | 16000 |
|
||||
| Outputs | AUC |
|
||||
|
|
|
|||
|
|
@ -16,17 +16,20 @@
|
|||
- [生成和处理合成数据](#生成和处理合成数据)
|
||||
- [训练过程](#训练过程)
|
||||
- [单机训练](#单机训练)
|
||||
- [单机训练缓存模式](#单机训练缓存模式)
|
||||
- [分布式训练](#分布式训练)
|
||||
- [参数服务器](#参数服务器)
|
||||
- [评估过程](#评估过程)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [推理过程](#推理过程)
|
||||
- [导出MindIR](#导出mindir)
|
||||
- [在Ascend310执行推理](#在ascend310执行推理)
|
||||
- [结果](#结果)
|
||||
- [result](#result)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [极致性能体验](#极致性能体验)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
|
|
@ -455,8 +458,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATASET_PATH] [DATA_TYPE] [NEED_PREPROCESS]
|
|||
| 参数 | Ascend单机 | GPU单机 | 数据并行模式-8卡 | 主机设备模式-8卡 |
|
||||
| ------------------------ | ------------------------------- | ------------------------------- | ------------------------------- | ------------------------------- |
|
||||
| 资源 |Ascend 910；系统 Euler2.8 | Tesla V100-PCIE 32G | Ascend 910；系统 Euler2.8 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-08-21 | 2020-08-21 | 2020-08-21 | 2020-08-21 |
|
||||
| MindSpore版本 | 0.6.0-beta | master | 0.6.0-beta | 0.6.0-beta |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | [1] | [1] | [1] | [1] |
|
||||
| 训练参数 | Epoch=15,<br />batch_size=16000 | Epoch=15,<br />batch_size=16000 | Epoch=15,<br />batch_size=16000 | Epoch=15,<br />batch_size=16000 |
|
||||
| 优化器 | FTRL,Adam | FTRL,Adam | FTRL,Adam | FTRL,Adam |
|
||||
|
|
@ -476,8 +479,8 @@ bash run_infer_310.sh [MINDIR_PATH] [DATASET_PATH] [DATA_TYPE] [NEED_PREPROCESS]
|
|||
| 参数 | Wide&Deep |
|
||||
| ----------------- | --------------------------- |
|
||||
| 资源 | Ascend 910；系统 Euler2.8 |
|
||||
| 上传日期 | 2020-08-21 |
|
||||
| MindSpore版本 | 0.6.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 |
|
||||
| 数据集 | [1] |
|
||||
| 批次大小 | 16000 |
|
||||
| 输出 | AUC |
|
||||
|
|
|
|||
|
|
@ -1,24 +1,25 @@
|
|||
# Contents
|
||||
|
||||
- [Wide&Deep Description](#widedeep-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Script Parameters](#training-script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [SingleDevice](#singledevice)
|
||||
- [Distribute Training](#distribute-training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [Contents](#contents)
|
||||
- [Wide&Deep Description](#widedeep-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Script Parameters](#training-script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [SingleDevice](#singledevice)
|
||||
- [Distribute Training](#distribute-training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## [Wide&Deep Description](#contents)
|
||||
|
||||
|
|
@ -168,8 +169,8 @@ python eval.py
|
|||
| Parameters | Single <br />Ascend | Data-Parallel-8P |
|
||||
| ------------------------ | ------------------------------- | ------------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 | Ascend 910 |
|
||||
| Uploaded Date | 08/21/2020 (month/day/year) | 08/21/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0 | 1.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | [1] | [1] |
|
||||
| Training Parameters | Epoch=3,<br />batch_size=131072 | Epoch=8,<br />batch_size=131072 |
|
||||
| Optimizer | FTRL,Adam | FTRL,Adam |
|
||||
|
|
@ -188,8 +189,8 @@ All executable scripts can be found in [here](https://gitee.com/mindspore/mindsp
|
|||
| Parameters | Wide&Deep |
|
||||
| ----------------- | --------------------------- |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 10/27/2020 (month/day/year) |
|
||||
| MindSpore Version | 1.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | [1] |
|
||||
| Batch Size | 131072 |
|
||||
| Outputs | AUC，MAP |
|
||||
|
|
|
|||
|
|
@ -1,11 +1,11 @@
|
|||
# 目录
|
||||
|
||||
- [目录](#目录)
|
||||
- [Wide&Deep概述](#widedeep概述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [Wide&Deep概述](#widedeep概述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
- [快速入门](#快速入门)
|
||||
- [脚本说明](#脚本说明)
|
||||
- [脚本和样例代码](#脚本和样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
|
|
@ -14,10 +14,10 @@
|
|||
- [单机训练](#单机训练)
|
||||
- [分布训练](#分布训练)
|
||||
- [评估过程](#评估过程)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [评估性能](#评估性能)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
|
|
@ -170,8 +170,8 @@ python eval.py
|
|||
| 参数 | 单Ascend | 数据并行-8卡 |
|
||||
| ------------------------ | ------------------------------- | ------------------------------- |
|
||||
| 资源 | Ascend 910 | Ascend 910 |
|
||||
| 上传日期 | 2020-08-21 | 2020-08-21 |
|
||||
| MindSpore版本 | 0.7.0-beta | 0.7.0-beta |
|
||||
| 上传日期 | 2021-07-05 | 2021-07-05 |
|
||||
| MindSpore版本 | 1.3.0 | 1.3.0 |
|
||||
| 数据集 | [1] | [1] |
|
||||
| 训练参数 | Epoch=3,<br />batch_size=131072 | Epoch=8,<br />batch_size=131072 |
|
||||
| 优化器 | FTRL,Adam | FTRL,Adam |
|
||||
|
|
@ -190,8 +190,8 @@ python eval.py
|
|||
| 参数 | Wide&Deep |
|
||||
| ----------------- | --------------------------- |
|
||||
| 资源 | Ascend 910 |
|
||||
| 上传日期 | 2020-08-21 |
|
||||
| MindSpore 版本 | 0.7.0-beta |
|
||||
| 上传日期 | 2021-07-05 |
|
||||
| MindSpore 版本 | 1.3.0 |
|
||||
| 数据集 | [1] |
|
||||
| 批次大小 | 131072 |
|
||||
| 输出 | AUC，MAP |
|
||||
|
|
|
|||
|
|
@ -1,26 +1,31 @@
|
|||
# Contents
|
||||
|
||||
- [FCN-4 Description](#fcn-4-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Contents](#contents)
|
||||
- [FCN-4 Description](#fcn-4-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [1. Download and preprocess the dataset](#1-download-and-preprocess-the-dataset)
|
||||
- [2. setup parameters (src/model_utils/default_config.yaml)](#2-setup-parameters-srcmodel_utilsdefault_configyaml)
|
||||
- [3. Train](#3-train)
|
||||
- [4. Test](#4-test)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
## [FCN-4 Description](#contents)
|
||||
|
||||
|
|
@ -309,8 +314,8 @@ AUC: 0.90995
|
|||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| Model Version | FCN-4 |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 56cores; Memory 314G; OS Euler2.8 |
|
||||
| uploaded Date | 09/11/2020 (month/day/year) |
|
||||
| MindSpore Version | r0.7.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Training Parameters | epoch=10, steps=534, batch_size = 32, lr=0.005 |
|
||||
| Optimizer | Adam |
|
||||
| Loss Function | Binary cross entropy |
|
||||
|
|
|
|||
|
|
@ -1,19 +1,22 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [Manifold Dynamic Pruning Description](#manifold-dynamic-pruning-description)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision(Ascend))
|
||||
- [Mixed Precision(Ascend)](#mixed-precisionascend)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Training Process](#training-process)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Script description](#script-description)
|
||||
- [Script and sample code](#script-and-sample-code)
|
||||
- [Training process](#training-process)
|
||||
- [Eval process](#eval-process)
|
||||
- [Usage](#usage)
|
||||
- [Launch](#launch)
|
||||
- [Result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#evaluation-performance)
|
||||
- [Inference Performance](#evaluation-performance)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [ResNet20 on CIFAR-10](#resnet20-on-cifar-10)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -105,7 +108,7 @@ result: {'acc': 0.9204727564102564}
|
|||
| -------------------------- | -------------------------------------- |
|
||||
| Model Version | ResNet20 |
|
||||
| uploaded Date | 03/27/2021 (month/day/year) ； |
|
||||
| MindSpore Version | 0.6.0-alpha |
|
||||
| MindSpore Version | 1.2.1 |
|
||||
| Dataset | CIFAR-10 |
|
||||
| Parameters (M) | 0.27 |
|
||||
| FLOPs (M) | 18.74 |
|
||||
|
|
|
|||
|
|
@ -1,6 +1,9 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [SE-Net Description](#se-net-description)
|
||||
- [Description](#description)
|
||||
- [Paper](#paper)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
|
|
@ -11,16 +14,27 @@
|
|||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Usage](#usage)
|
||||
- [Running on Ascend](#running-on-ascend)
|
||||
- [Running on GPU](#running-on-gpu)
|
||||
- [Result](#result)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Usage](#usage-1)
|
||||
- [Running on Ascend](#running-on-ascend-1)
|
||||
- [Running on GPU](#running-on-gpu-1)
|
||||
- [Result](#result-1)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result-2)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [SE-ResNet50 on ImageNet2012](#se-resnet50-on-imagenet2012)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [SE-ResNet50 on ImageNet2012](#se-resnet50-on-imagenet2012-1)
|
||||
- [310Inference Performance](#310inference-performance)
|
||||
- [SE-ResNet50 on ImageNet2012](#se-resnet50-on-imagenet2012-2)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -284,8 +298,8 @@ result: {'top_5_accuracy': 93.86%, 'top_1_accuracy': 77.80%}
|
|||
| -------------------------- | ---------------------------------------------------------- | ---------------------------------------------------------- |
|
||||
| Model Version | SE-ResNet50 | SE-ResNet50 |
|
||||
| Resource | CentOs 8.2, Ascend 910，CPU 2.60GHz 192cores，Memory 755G | V100-PCIE 32G |
|
||||
| uploaded Date | 03/19/2021 (month/day/year) | 07/14/2021 (month/day/year) |
|
||||
| MindSpore Version | 0.7.0-alpha | 1.3.0 |
|
||||
| uploaded Date | 07/05/2021 (month/day/year) | 07/14/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet2012 | ImageNet2012 |
|
||||
| Training Parameters | epoch=90, steps per epoch=5004, batch_size = 256 | epoch=90, steps per epoch=5004, batch_size = 256 |
|
||||
| Optimizer | Momentum | Momentum |
|
||||
|
|
@ -306,8 +320,8 @@ result: {'top_5_accuracy': 93.86%, 'top_1_accuracy': 77.80%}
|
|||
| ------------------- | --------------------------- | --------------------------- |
|
||||
| Model Version | SE-ResNet50 | SE-ResNet50 |
|
||||
| Resource | Ascend 910 | V100-PCIE 32G |
|
||||
| Uploaded Date | 03/19/2021 (month/day/year) | 07/14/2021 (month/day/year) |
|
||||
| MindSpore Version | 0.7.0-alpha | 1.3.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) | 07/14/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 | 1.3.0 |
|
||||
| Dataset | ImageNet2012 | ImageNet2012 |
|
||||
| batch_size | 256 | 256 |
|
||||
| Accuracy | 77.74% | 77.66% |
|
||||
|
|
|
|||
|
|
@ -3,7 +3,10 @@
|
|||
|
||||
<!-- TOC -->
|
||||
|
||||
- [Glore_resnet200描述](#Glore_resnet200描述)
|
||||
- [目录](#目录)
|
||||
- [Glore_resnet200描述](#glore_resnet200描述)
|
||||
- [概述](#概述)
|
||||
- [论文](#论文)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [特性](#特性)
|
||||
|
|
@ -14,19 +17,23 @@
|
|||
- [脚本及样例代码](#脚本及样例代码)
|
||||
- [脚本参数](#脚本参数)
|
||||
- [训练过程](#训练过程)
|
||||
- [用法](#用法)
|
||||
- [Ascend处理器环境运行](#ascend处理器环境运行)
|
||||
- [GPU处理器环境运行](#gpu处理器环境运行)
|
||||
- [训练结果](#训练结果)
|
||||
- [推理过程](#推理过程)
|
||||
- [用法](#用法-1)
|
||||
- [Ascend处理器环境运行](#ascend处理器环境运行-1)
|
||||
- [GPU处理器环境运行](#gpu处理器环境运行-1)
|
||||
- [推理结果](#推理结果)
|
||||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [Imagenet2012上的Glore_resnet200](#Imagenet2012上的Glore_resnet200)
|
||||
- [ImageNet2012上的Glore_resnet200](#imagenet2012上的glore_resnet200)
|
||||
- [推理性能](#推理性能)
|
||||
- [Imagenet2012上的Glore_resnet200](#Imagenet2012上的Glore_resnet200)
|
||||
- [使用流程](#使用流程)
|
||||
- [推理](#推理)
|
||||
- [ImageNet2012上的Glore_resnet200](#imagenet2012上的glore_resnet200-1)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#ModelZoo主页)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
<!-- /TOC -->
|
||||
|
||||
|
|
@ -285,7 +292,7 @@ result:{'top_1 acc':0.802303685897436}
|
|||
| 模型版本 | Glore_resnet200 |Glore_resnet200 |
|
||||
| 资源 | Ascend 910；CPU：2.60GHz，192核；内存：2048G |GPU-V100(SXM2) |
|
||||
| 上传日期 | 2021-03-34 |2021-05-25 |
|
||||
| MindSpore版本 | 1.1.1-c76 |1.2.0-rc1 |
|
||||
| MindSpore版本 | 1.1.1 |1.2.0 |
|
||||
| 数据集 | ImageNet2012 | ImageNet2012 |
|
||||
| 训练参数 | epoch=150, steps per epoch=1251, batch_size = 128 |epoch=150, steps per epoch=2502, batch_size = 64 |
|
||||
| 优化器 | NAG | NAG |
|
||||
|
|
@ -307,7 +314,7 @@ result:{'top_1 acc':0.802303685897436}
|
|||
| 模型版本 | Glore_resnet200 | Glore_resnet200 |
|
||||
| 资源 | Ascend 910 | GPU |
|
||||
| 上传日期 | 2021-3-24 |2021-05-25 |
|
||||
| MindSpore版本 | 1.1.1-c76 |1.2.0-rc1 |
|
||||
| MindSpore版本 | 1.1.1 |1.2.0 |
|
||||
| 数据集 | 12万张图像 |12万张图像 |
|
||||
| batch_size | 128 |64 |
|
||||
| 输出 | 概率 |概率 |
|
||||
|
|
|
|||
|
|
@ -1,19 +1,22 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [ReNAS Description](#renas-description)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Mixed Precision(Ascend)](#mixed-precisionascend)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Training Process](#training-process)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Script description](#script-description)
|
||||
- [Script and sample code](#script-and-sample-code)
|
||||
- [Training process](#training-process)
|
||||
- [Eval process](#eval-process)
|
||||
- [Usage](#usage)
|
||||
- [Launch](#launch)
|
||||
- [Result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#evaluation-performance)
|
||||
- [Inference Performance](#evaluation-performance)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [NASBench101-Net on CIFAR-10](#nasbench101-net-on-cifar-10)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -104,7 +107,7 @@ result: {'acc': 0.9411057692307693} ckpt= ./resnet50-imgnet-0.65x-80.24.ckpt
|
|||
| -------------------------- | -------------------------------------- |
|
||||
| Model Version | NASBench101-Net |
|
||||
| uploaded Date | 03/27/2021 (month/day/year) ； |
|
||||
| MindSpore Version | 0.6.0-alpha |
|
||||
| MindSpore Version | 1.2.0 |
|
||||
| Dataset | CIFAR-10 |
|
||||
| Parameters (M) | 4.44 |
|
||||
| FLOPs (G) | 1.9 |
|
||||
|
|
|
|||
|
|
@ -1,26 +1,26 @@
|
|||
# Contents
|
||||
|
||||
- [Contents](#contents)
|
||||
- [SSD Description](#ssd-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training on Ascend](#training-on-ascend)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation on Ascend](#evaluation-on-ascend)
|
||||
- [SSD Description](#ssd-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Training Process](#training-process)
|
||||
- [Training on Ascend](#training-on-ascend)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation on Ascend](#evaluation-on-ascend)
|
||||
- [Inference Process](#inference-process)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Export MindIR](#export-mindir)
|
||||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [310Inference Performance](#310inference-performance)
|
||||
|
||||
# [SSD Description](#contents)
|
||||
|
|
@ -312,7 +312,7 @@ mAP: 0.24270569394180577
|
|||
| -------------------------- | -------------------------------------------------------------|
|
||||
| Model Version | SSD ghostnet |
|
||||
| Resource | Ascend 910; CPU 2.60GHz, 192cores; Memory 755G; OS Euler2.8 |
|
||||
| MindSpore Version | 0.7.0 |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | COCO2017 |
|
||||
| Training Parameters | epoch = 500, batch_size = 32 |
|
||||
| Optimizer | Momentum |
|
||||
|
|
@ -325,8 +325,8 @@ mAP: 0.24270569394180577
|
|||
| ------------------- | ----------------------------|
|
||||
| Model Version | SSD ghostnet |
|
||||
| Resource | Ascend 910; OS Euler2.8 |
|
||||
| Uploaded Date | 09/08/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.7.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | COCO2017 |
|
||||
| batch_size | 1 |
|
||||
| outputs | mAP |
|
||||
|
|
@ -339,8 +339,8 @@ mAP: 0.24270569394180577
|
|||
| ------------------- | --------------------------- |
|
||||
| Model Version | SSD ghostnet |
|
||||
| Resource | Ascend 310; OS Euler2.8 |
|
||||
| Uploaded Date | 06/01/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.2.0 |
|
||||
| Uploaded Date | 07/05/2021 (month/day/year) |
|
||||
| MindSpore Version | 1.3.0 |
|
||||
| Dataset | COCO2017 |
|
||||
| batch_size | 1 |
|
||||
| outputs | mAP |
|
||||
|
|
|
|||
|
|
@ -3,7 +3,7 @@
|
|||
<!-- TOC -->
|
||||
|
||||
- [目录](#目录)
|
||||
- [WGAN描述](#WGAN描述)
|
||||
- [WGAN描述](#wgan描述)
|
||||
- [模型架构](#模型架构)
|
||||
- [数据集](#数据集)
|
||||
- [环境要求](#环境要求)
|
||||
|
|
@ -22,9 +22,10 @@
|
|||
- [模型描述](#模型描述)
|
||||
- [性能](#性能)
|
||||
- [训练性能](#训练性能)
|
||||
- [第一种情况（选用标准卷积DCGAN的生成器结构）](#第一种情况（选用标准卷积DCGAN的生成器结构）)
|
||||
- [第二种情况（选用没有BatchNorm的卷积DCGAN的生成器结构）](#第二种情况（选用没有BatchNorm的卷积DCGAN的生成器结构）)
|
||||
- [第一种情况（选用标准卷积DCGAN的生成器结构）](#第一种情况选用标准卷积dcgan的生成器结构)
|
||||
- [第二种情况（选用没有BatchNorm的卷积DCGAN的生成器结构）](#第二种情况选用没有batchnorm的卷积dcgan的生成器结构)
|
||||
- [推理性能](#推理性能)
|
||||
- [推理](#推理-1)
|
||||
- [随机情况说明](#随机情况说明)
|
||||
- [ModelZoo主页](#modelzoo主页)
|
||||
|
||||
|
|
@ -238,7 +239,7 @@ bash run_infer_310.sh [MINDIR_PATH] [CONFIG_PATH] [NEED_PREPROCESS] [NIMAGES] [D
|
|||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| 资源 | Ascend 910 ；CPU 2.60GHz，192核；内存：755G |
|
||||
| 上传日期 | 2021-05-14 |
|
||||
| MindSpore版本 | 1.2.0-alpha |
|
||||
| MindSpore版本 | 1.2.0 |
|
||||
| 数据集 | LSUN-Bedrooms |
|
||||
| 训练参数 | max_epoch=25, batch_size=64, lr_init=0.00005 |
|
||||
| 优化器 | RMSProp |
|
||||
|
|
@ -261,7 +262,7 @@ bash run_infer_310.sh [MINDIR_PATH] [CONFIG_PATH] [NEED_PREPROCESS] [NIMAGES] [D
|
|||
| -------------------------- | ----------------------------------------------------------- |
|
||||
| 资源 | Ascend 910 ；CPU 2.60GHz，192核；内存：755G |
|
||||
| 上传日期 | 2021-05-14 |
|
||||
| MindSpore版本 | 1.2.0-alpha |
|
||||
| MindSpore版本 | 1.2.0 |
|
||||
| 数据集 | LSUN-Bedrooms |
|
||||
| 训练参数 | max_epoch=25, batch_size=64, lr_init=0.00005 |
|
||||
| 优化器 | RMSProp |
|
||||
|
|
@ -286,7 +287,7 @@ bash run_infer_310.sh [MINDIR_PATH] [CONFIG_PATH] [NEED_PREPROCESS] [NIMAGES] [D
|
|||
| ------------------- | --------------------------- |
|
||||
| 资源 | Ascend 910 |
|
||||
| 上传日期 | 2021-05-14 |
|
||||
| MindSpore 版本 | 1.2.0-alpha |
|
||||
| MindSpore 版本 | 1.2.0 |
|
||||
| 数据集 | LSUN-Bedrooms |
|
||||
| batch_size | 1 |
|
||||
| 输出 | 生成的图片 |
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
# Contents
|
||||
|
||||
- [AutoDis Description](#AutoDis-description)
|
||||
- [Contents](#contents)
|
||||
- [AutoDis Description](#autodis-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
|
|
@ -17,9 +18,9 @@
|
|||
- [Infer on Ascend310](#infer-on-ascend310)
|
||||
- [result](#result)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Inference Performance](#evaluation-performance)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Inference Performance](#inference-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
|
|
@ -312,7 +313,7 @@ Inference result is saved in current path, you can find result in acc.log file.
|
|||
| Model Version | AutoDis |
|
||||
| Resource | Ascend 910 |
|
||||
| Uploaded Date | 12/12/2020 (month/day/year) |
|
||||
| MindSpore Version | 0.3.0-alpha |
|
||||
| MindSpore Version | 1.1.0 |
|
||||
| Dataset | [1] |
|
||||
| batch_size | 1000 |
|
||||
| outputs | accuracy |
|
||||
|
|
|
|||
Loading…
Reference in New Issue