diff --git a/docs/federated/docs/source_en/index.rst b/docs/federated/docs/source_en/index.rst index 9ba4357203e578545cc36b4adc655f2fd4fbd3b4..ed6773c901f2bc0afa3a39229f7cb51ca933e363 100644 --- a/docs/federated/docs/source_en/index.rst +++ b/docs/federated/docs/source_en/index.rst @@ -12,7 +12,7 @@ The federated learning is an encrypted distributed machine learning technology t .. raw:: html - + Advantages of the MindSpore Federated --------------------------------------- diff --git a/docs/federated/docs/source_zh_cn/object_detection_application_in_cross_silo.md b/docs/federated/docs/source_zh_cn/object_detection_application_in_cross_silo.md index 168b8120fd8e796271aff7f59f1ab5d1ab359edf..792deb8e73c335a2e54eb79209459835dfea55a6 100644 --- a/docs/federated/docs/source_zh_cn/object_detection_application_in_cross_silo.md +++ b/docs/federated/docs/source_zh_cn/object_detection_application_in_cross_silo.md @@ -8,7 +8,7 @@ ## 任务前准备 -本教程基于MindSpore model_zoo中提供的的faster_rcnn网络部署云云联邦目标检测任务，请先根据官方[faster_rcnn教程及代码](https://gitee.com/mindspore/models/tree/master/official/cv/faster_rcnn)先了解COCO数据集、faster_rcnn网络结构、训练过程以及评估过程。由于COCO数据集已开源，请参照其[官网](https://cocodataset.org/#home)指引自行下载好数据集，并进行数据集切分（例如模拟100个客户端，可将数据集切分成100份，每份代表一个客户端所持有的数据）。 +本教程基于MindSpore model_zoo中提供的的faster_rcnn网络部署云云联邦目标检测任务，请先根据官方[faster_rcnn教程及代码](https://gitee.com/mindspore/models/tree/r1.7/official/cv/faster_rcnn)先了解COCO数据集、faster_rcnn网络结构、训练过程以及评估过程。由于COCO数据集已开源，请参照其[官网](https://cocodataset.org/#home)指引自行下载好数据集，并进行数据集切分（例如模拟100个客户端，可将数据集切分成100份，每份代表一个客户端所持有的数据）。 由于原始COCO数据集为json文件格式，云云联邦学习框架提供的目标检测脚本暂时只支持MindRecord格式输入数据，可根据以下步骤将json文件转换为MindRecord格式文件： @@ -96,7 +96,7 @@ cross_silo_faster_rcnn 用于设置预训练模型路径（.ckpt 格式） - 本教程中实验的预训练模型是在ImageNet2012上训练的ResNet-50检查点。你可以使用ModelZoo中 [resnet50](https://gitee.com/mindspore/models/tree/master/official/cv/resnet) 脚本来训练，然后使用src/convert_checkpoint.py把训练好的resnet50的权重文件转换为可加载的权重文件。 + 本教程中实验的预训练模型是在ImageNet2012上训练的ResNet-50检查点。你可以使用ModelZoo中 [resnet50](https://gitee.com/mindspore/models/tree/r1.7/official/cv/resnet) 脚本来训练，然后使用src/convert_checkpoint.py把训练好的resnet50的权重文件转换为可加载的权重文件。 3. 启动Scheduler diff --git a/docs/lite/docs/source_en/image_segmentation_lite.md b/docs/lite/docs/source_en/image_segmentation_lite.md index a456a59e3d5438918fe286b1efb0dee74d5388de..3e08e4e779546f74dc44c656b22552e80f4cb040 100644 --- a/docs/lite/docs/source_en/image_segmentation_lite.md +++ b/docs/lite/docs/source_en/image_segmentation_lite.md @@ -6,7 +6,7 @@ Image segmentation is used to detect the position of the object in the picture or a pixel belongs to which object. -Using MindSpore Lite to perform image segmentation [example](https://gitee.com/mindspore/models/tree/master/official/lite/image_segmentation). +Using MindSpore Lite to perform image segmentation [example](https://gitee.com/mindspore/models/tree/r1.7/official/lite/image_segmentation). ## Image segmentation model list diff --git a/docs/lite/docs/source_en/object_detection_lite.md b/docs/lite/docs/source_en/object_detection_lite.md index b8ff787884637aa0e6a22ffb2bedf59717e643fc..8335d8fd87c8943a783544c56faa8b12cae49521 100644 --- a/docs/lite/docs/source_en/object_detection_lite.md +++ b/docs/lite/docs/source_en/object_detection_lite.md @@ -12,7 +12,7 @@ Object detection can identify the object in the image and its position in the im | -------- | ----------- | ---------------- | | mouse | 0.78 | [10, 25, 35, 43] | -Using MindSpore Lite to implement object detection [example](https://gitee.com/mindspore/models/tree/master/official/lite/object_detection). +Using MindSpore Lite to implement object detection [example](https://gitee.com/mindspore/models/tree/r1.7/official/lite/object_detection). ## Object detection model list diff --git a/docs/lite/docs/source_en/quick_start/image_segmentation.md b/docs/lite/docs/source_en/quick_start/image_segmentation.md index fae0b690ea80f95f5d071a5a9236e919b392c7a9..1dfd76508f110b274e2041b4b0950aaddf515e20 100644 --- a/docs/lite/docs/source_en/quick_start/image_segmentation.md +++ b/docs/lite/docs/source_en/quick_start/image_segmentation.md @@ -34,7 +34,7 @@ The following describes how to build and execute an on-device image segmentation ### Building and Running -1. Load the [sample source code](https://gitee.com/mindspore/models/tree/master/official/lite/image_segmentation) to Android Studio and install the corresponding SDK. (After the SDK version is specified, Android Studio automatically installs the SDK.) +1. Load the [sample source code](https://gitee.com/mindspore/models/tree/r1.7/official/lite/image_segmentation) to Android Studio and install the corresponding SDK. (After the SDK version is specified, Android Studio automatically installs the SDK.) ![start_home](../images/lite_quick_start_home.png) @@ -117,7 +117,7 @@ Note: If the download fails, manually download the model file [segment_model.ms] ### Writing On-Device Inference Code -The inference code and process are as follows. For details about the complete code, see [src/java/com/mindspore/imagesegmentation/TrackingMobile](https://gitee.com/mindspore/models/blob/master/official/lite/image_segmentation/app/src/main/java/com/mindspore/imagesegmentation/help/TrackingMobile.java). +The inference code and process are as follows. For details about the complete code, see [src/java/com/mindspore/imagesegmentation/TrackingMobile](https://gitee.com/mindspore/models/blob/r1.7/official/lite/image_segmentation/app/src/main/java/com/mindspore/imagesegmentation/help/TrackingMobile.java). 1. Load the MindSpore Lite model file and build the context, model, and computational graph for inference. diff --git a/docs/lite/docs/source_en/quick_start/train_lenet.md b/docs/lite/docs/source_en/quick_start/train_lenet.md index 6147e6552b710af4f8e5ea4c4c60082b95b70036..5981767559c0aaad152cc287b6252ed368a01d68 100644 --- a/docs/lite/docs/source_en/quick_start/train_lenet.md +++ b/docs/lite/docs/source_en/quick_start/train_lenet.md @@ -213,7 +213,7 @@ train_lenet_cpp/ ### Model Exporting -Whether it is an off-the-shelf prepared model, or a custom written model, the model needs to be exported to a `.mindir` file. Here we use the already-implemented [LeNet model](https://gitee.com/mindspore/models/tree/master/official/cv/lenet). +Whether it is an off-the-shelf prepared model, or a custom written model, the model needs to be exported to a `.mindir` file. Here we use the already-implemented [LeNet model](https://gitee.com/mindspore/models/tree/r1.7/official/cv/lenet). Import and instantiate a LeNet5 model and set the model to train mode: diff --git a/docs/lite/docs/source_en/scene_detection_lite.md b/docs/lite/docs/source_en/scene_detection_lite.md index c1df909ceef4fbacefdd08f939a57bf44364e0c5..4eb468c1916c923366863c6241012dc815983272 100644 --- a/docs/lite/docs/source_en/scene_detection_lite.md +++ b/docs/lite/docs/source_en/scene_detection_lite.md @@ -6,7 +6,7 @@ Scene detection can identify the type of scene in the device's camera. -Using MindSpore Lite to implement scene detection [example](https://gitee.com/mindspore/models/tree/master/official/lite/scene_detection). +Using MindSpore Lite to implement scene detection [example](https://gitee.com/mindspore/models/tree/r1.7/official/lite/scene_detection). ## Scene detection model list diff --git a/docs/lite/docs/source_en/style_transfer_lite.md b/docs/lite/docs/source_en/style_transfer_lite.md index 71eadeb3dcf6bcb67e59ee39d86218517fd57113..34a5896f2084c83156e5fbbb0b75b4d9ae954117 100644 --- a/docs/lite/docs/source_en/style_transfer_lite.md +++ b/docs/lite/docs/source_en/style_transfer_lite.md @@ -14,4 +14,4 @@ Selecting the first standard image from left to perform the style transfer, as s ![image_after_transfer](images/after_transfer.png) -Using MindSpore Lite to realize style transfer [example](https://gitee.com/mindspore/models/tree/master/official/lite/style_transfer). +Using MindSpore Lite to realize style transfer [example](https://gitee.com/mindspore/models/tree/r1.7/official/lite/style_transfer). diff --git a/docs/lite/docs/source_zh_cn/image_segmentation_lite.md b/docs/lite/docs/source_zh_cn/image_segmentation_lite.md index 634af311299d2838196671605a0cf9db0bab2fa8..974185c666c68db24b683a9b97af49f3cb516307 100644 --- a/docs/lite/docs/source_zh_cn/image_segmentation_lite.md +++ b/docs/lite/docs/source_zh_cn/image_segmentation_lite.md @@ -6,7 +6,7 @@ 图像分割是用于检测目标在图片中的位置或者图片中某一像素是输入何种对象的。 -使用MindSpore Lite实现图像分割的[示例代码](https://gitee.com/mindspore/models/tree/master/official/lite/image_segmentation)。 +使用MindSpore Lite实现图像分割的[示例代码](https://gitee.com/mindspore/models/tree/r1.7/official/lite/image_segmentation)。 ## 图像分割模型列表 diff --git a/docs/lite/docs/source_zh_cn/object_detection_lite.md b/docs/lite/docs/source_zh_cn/object_detection_lite.md index 0f47b861a460777145fe2fcc1b4f60267d410207..edf9605e6c52eb9b705dca8b9c1d889ebb72ec74 100644 --- a/docs/lite/docs/source_zh_cn/object_detection_lite.md +++ b/docs/lite/docs/source_zh_cn/object_detection_lite.md @@ -12,7 +12,7 @@ | ----- | ---- | ---------------- | | mouse | 0.78 | [10, 25, 35, 43] | -使用MindSpore Lite实现目标检测的[示例代码](https://gitee.com/mindspore/models/tree/master/official/lite/object_detection)。 +使用MindSpore Lite实现目标检测的[示例代码](https://gitee.com/mindspore/models/tree/r1.7/official/lite/object_detection)。 ## 目标检测模型列表 diff --git a/docs/lite/docs/source_zh_cn/quick_start/image_segmentation.md b/docs/lite/docs/source_zh_cn/quick_start/image_segmentation.md index eeac5cb055388a9b7ceefbd90a0b9115b5fab538..4404e289efec98284e93a831b8d728168b89d8da 100644 --- a/docs/lite/docs/source_zh_cn/quick_start/image_segmentation.md +++ b/docs/lite/docs/source_zh_cn/quick_start/image_segmentation.md @@ -34,7 +34,7 @@ ### 构建与运行 -1. 在Android Studio中加载本[示例源码](https://gitee.com/mindspore/models/tree/master/official/lite/image_segmentation)，并安装相应的SDK（指定SDK版本后，由Android Studio自动安装）。 +1. 在Android Studio中加载本[示例源码](https://gitee.com/mindspore/models/tree/r1.7/official/lite/image_segmentation)，并安装相应的SDK（指定SDK版本后，由Android Studio自动安装）。 ![start_home](../images/lite_quick_start_home.png) @@ -117,7 +117,7 @@ Android调用MindSpore Android AAR时，需要相关库文件支持。可通过M ### 编写端侧推理代码 -推理代码流程如下，完整代码请参见 [src/java/com/mindspore/imagesegmentation/TrackingMobile](https://gitee.com/mindspore/models/blob/master/official/lite/image_segmentation/app/src/main/java/com/mindspore/imagesegmentation/help/TrackingMobile.java)。 +推理代码流程如下，完整代码请参见 [src/java/com/mindspore/imagesegmentation/TrackingMobile](https://gitee.com/mindspore/models/blob/r1.7/official/lite/image_segmentation/app/src/main/java/com/mindspore/imagesegmentation/help/TrackingMobile.java)。 1. 加载MindSpore Lite模型，构建上下文、会话以及用于推理的计算图。 diff --git a/docs/lite/docs/source_zh_cn/quick_start/train_lenet.md b/docs/lite/docs/source_zh_cn/quick_start/train_lenet.md index 4f0298017d35f1d48005a9646168ef518478ecc6..b3dc066802d1f6d5e2a0537929a9ecf194c1f122 100644 --- a/docs/lite/docs/source_zh_cn/quick_start/train_lenet.md +++ b/docs/lite/docs/source_zh_cn/quick_start/train_lenet.md @@ -212,7 +212,7 @@ The predicted classes are: ### 定义并导出模型 -首先我们需要基于MindSpore框架创建一个LeNet模型，本例中直接用MindSpore ModelZoo的现有[LeNet模型](https://gitee.com/mindspore/models/tree/master/official/cv/lenet)。 +首先我们需要基于MindSpore框架创建一个LeNet模型，本例中直接用MindSpore ModelZoo的现有[LeNet模型](https://gitee.com/mindspore/models/tree/r1.7/official/cv/lenet)。 > 本小结使用MindSpore云侧功能导出，更多信息请参考[MindSpore教程](https://www.mindspore.cn/tutorials/experts/zh-CN/r1.7/index.html)。 diff --git a/docs/lite/docs/source_zh_cn/scene_detection_lite.md b/docs/lite/docs/source_zh_cn/scene_detection_lite.md index 5056f72e2aa477639292b9218843a57104077082..9b8a13464a804b9364d20f0ffafd2806832e942d 100644 --- a/docs/lite/docs/source_zh_cn/scene_detection_lite.md +++ b/docs/lite/docs/source_zh_cn/scene_detection_lite.md @@ -6,7 +6,7 @@ 场景检测可以识别设备摄像头中场景的类型。 -使用MindSpore Lite实现场景检测的[示例代码](https://gitee.com/mindspore/models/tree/master/official/lite/scene_detection)。 +使用MindSpore Lite实现场景检测的[示例代码](https://gitee.com/mindspore/models/tree/r1.7/official/lite/scene_detection)。 ## 场景检测模型列表 diff --git a/docs/lite/docs/source_zh_cn/style_transfer_lite.md b/docs/lite/docs/source_zh_cn/style_transfer_lite.md index e39366dad4ce43edb0d21675eca44d9f55de3aa7..673c845584fe6479399e9032ed3160c41615daf6 100644 --- a/docs/lite/docs/source_zh_cn/style_transfer_lite.md +++ b/docs/lite/docs/source_zh_cn/style_transfer_lite.md @@ -14,4 +14,4 @@ ![image_after_transfer](images/after_transfer.png) -使用MindSpore Lite实现风格迁移的[示例代码](https://gitee.com/mindspore/models/tree/master/official/lite/style_transfer)。 +使用MindSpore Lite实现风格迁移的[示例代码](https://gitee.com/mindspore/models/tree/r1.7/official/lite/style_transfer)。 diff --git a/docs/mindarmour/docs/source_zh_cn/evaluation_of_CNNCTC.md b/docs/mindarmour/docs/source_zh_cn/evaluation_of_CNNCTC.md index 06aa323940086b679abcd1de42da320d3665dd1e..42573d522c3aa8f85434f7adcdd115503acc8c25 100644 --- a/docs/mindarmour/docs/source_zh_cn/evaluation_of_CNNCTC.md +++ b/docs/mindarmour/docs/source_zh_cn/evaluation_of_CNNCTC.md @@ -38,7 +38,7 @@ ### 脚本参数 -在`default_config.yaml`中可以同时配置训练参数、推理参数、鲁棒性评测参数。这里我们重点关注在评测过程中使用到的参数，以及需要用户配置的参数，其余参数说明参考[CNN-CTC教程](https://gitee.com/mindspore/models/tree/master/official/cv/cnnctc)。 +在`default_config.yaml`中可以同时配置训练参数、推理参数、鲁棒性评测参数。这里我们重点关注在评测过程中使用到的参数，以及需要用户配置的参数，其余参数说明参考[CNN-CTC教程](https://gitee.com/mindspore/models/tree/r1.7/official/cv/cnnctc)。 - `--TEST_DATASET_PATH`：测试数据集路。 - `--CHECKPOINT_PATH`：checkpoint路径。 @@ -51,7 +51,7 @@ [论文](https://arxiv.org/abs/1904.01906)： J. Baek, G. Kim, J. Lee, S. Park, D. Han, S. Yun, S. J. Oh, and H. Lee, “What is wrong with scene text recognition model comparisons? dataset and model analysis,” ArXiv, vol. abs/1904.01906, 2019. -数据处理与模型训练参考[CNN-CTC教程](https://gitee.com/mindspore/models/tree/master/official/cv/cnnctc)。评测任务需基于该教程获得预处理后的数据集和checkpoint模型文件。 +数据处理与模型训练参考[CNN-CTC教程](https://gitee.com/mindspore/models/tree/r1.7/official/cv/cnnctc)。评测任务需基于该教程获得预处理后的数据集和checkpoint模型文件。 预处理后的数据集为.lmdb格式，以键值对方式存储： diff --git a/docs/mindspore/source_en/design/cv_resnet50_second_order_optimizer.md b/docs/mindspore/source_en/design/cv_resnet50_second_order_optimizer.md index 8cdcd8d084b0d278c7118c680490b61b82618945..0b27d956e48c4e89d3624ef2e5cc087a7bec64f0 100644 --- a/docs/mindspore/source_en/design/cv_resnet50_second_order_optimizer.md +++ b/docs/mindspore/source_en/design/cv_resnet50_second_order_optimizer.md @@ -10,7 +10,7 @@ Based on the existing natural gradient algorithm, MindSpore development team use This tutorial describes how to use the second-order optimizer THOR provided by MindSpore to train the ResNet-50 v1.5 network and ImageNet dataset on Ascend 910 and GPU. > Download address of the complete code example: - + Directory Structure of Code Examples @@ -171,7 +171,7 @@ def create_dataset2(dataset_path, do_train, repeat_num=1, batch_size=32, target= ## Defining the Network -Use the ResNet-50 v1.5 network model as an example. Define the [ResNet-50 network](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/src/resnet.py). +Use the ResNet-50 v1.5 network model as an example. Define the [ResNet-50 network](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/src/resnet.py). After the network is built, call the defined ResNet-50 in the `__main__` function. @@ -334,7 +334,7 @@ bash run_distribute_train.sh Variables `RANK_TABLE_FILE`, `DATASET_PATH` and `CONFIG_PATH` need to be transferred to the script. The meanings of variables are as follows: -- `RANK_TABLE_FILE`: path for storing the networking information file (about the rank table file, you can refer to [HCCL_TOOL](https://gitee.com/mindspore/models/tree/master/utils/hccl_tools)) +- `RANK_TABLE_FILE`: path for storing the networking information file (about the rank table file, you can refer to [HCCL_TOOL](https://gitee.com/mindspore/models/tree/r1.7/utils/hccl_tools)) - `DATASET_PATH`: training dataset path - `CONFIG_PATH`: config file path diff --git a/docs/mindspore/source_en/design/host_device_training.md b/docs/mindspore/source_en/design/host_device_training.md index 25d45f2348605fed644a8379675b4a820ecf2020..35c215cc291764277d1fac28f31c8620c20c5d49 100644 --- a/docs/mindspore/source_en/design/host_device_training.md +++ b/docs/mindspore/source_en/design/host_device_training.md @@ -9,15 +9,15 @@ the number of required accelerators is too overwhelming for people to access, re efficient method for addressing huge model problem. In MindSpore, users can easily implement hybrid training by configuring trainable parameters and necessary operators to run on hosts, and other operators to run on accelerators. -This tutorial introduces how to train [Wide&Deep](https://gitee.com/mindspore/models/tree/master/official/recommend/wide_and_deep) in the Host+Ascend 910 AI Accelerator mode. +This tutorial introduces how to train [Wide&Deep](https://gitee.com/mindspore/models/tree/r1.7/official/recommend/wide_and_deep) in the Host+Ascend 910 AI Accelerator mode. ## Preliminaries -1. Prepare the model. The Wide&Deep code can be found at: , in which `train_and_eval_auto_parallel.py` is the main function for training, `src/` directory contains the model definition, data processing and configuration files, `script/` directory contains the launch scripts in different modes. +1. Prepare the model. The Wide&Deep code can be found at: , in which `train_and_eval_auto_parallel.py` is the main function for training, `src/` directory contains the model definition, data processing and configuration files, `script/` directory contains the launch scripts in different modes. 2. Prepare the dataset. Please refer the link in [1] to download the dataset, and use the script `src/preprocess_data.py` to transform dataset into MindRecord format. -3. Configure the device information. When performing training in the bare-metal environment, the network information file needs to be configured. This example only employs one accelerator, thus `rank_table_1p_0.json` containing #0 accelerator is configured (about the rank table file, you can refer to [HCCL_TOOL](https://gitee.com/mindspore/models/tree/master/utils/hccl_tools)). +3. Configure the device information. When performing training in the bare-metal environment, the network information file needs to be configured. This example only employs one accelerator, thus `rank_table_1p_0.json` containing #0 accelerator is configured (about the rank table file, you can refer to [HCCL_TOOL](https://gitee.com/mindspore/models/tree/r1.7/utils/hccl_tools)). ## Configuring for Hybrid Training diff --git a/docs/mindspore/source_en/design/parameter_server_training.md b/docs/mindspore/source_en/design/parameter_server_training.md index e00ac8afb1de52a06be91b1dfa816140c1e206b7..76991ce07897545df73155a23d9e0de29a182419 100644 --- a/docs/mindspore/source_en/design/parameter_server_training.md +++ b/docs/mindspore/source_en/design/parameter_server_training.md @@ -22,7 +22,7 @@ The following describes how to use parameter server to train LeNet on Ascend 910 ### Training Script Preparation -Learn how to train a LeNet using the [MNIST dataset](http://yann.lecun.com/exdb/mnist/) by referring to . +Learn how to train a LeNet using the [MNIST dataset](http://yann.lecun.com/exdb/mnist/) by referring to . ### Parameter Setting @@ -39,7 +39,7 @@ Learn how to train a LeNet using the [MNIST dataset](http://yann.lecun.com/exdb/ - The size of the weight which is updated by Parameter Server should not exceed INT_MAX(2^31 - 1) bytes. - The interface `set_param_ps` can receive a `bool` parameter:`init_in_server`, indicating whether this training parameter is initialized on the Server side. `init_in_server` defaults to `False`, indicating that this training parameter is initialized on Worker. Currently, only the training parameter `embedding_table` of the `EmbeddingLookup` operator is supported to be initialized on Server side to solve the problem of insufficient memory caused by the initialization of a large shape `embedding_table` on Worker. The `EmbeddingLookup` operator's `target` attribute needs to be set to 'CPU'. The training parameter initialized on the Server side will no longer be synchronized to Worker. If it involves multi-Server training and saves CheckPoint, each Server will save a CheckPoint after the training. -3. On the basis of the [original training script](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/train.py), set all LeNet model weights to be trained on the parameter server: +3. On the basis of the [original training script](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/train.py), set all LeNet model weights to be trained on the parameter server: ```python context.set_ps_context(enable_ps=True) @@ -47,7 +47,7 @@ Learn how to train a LeNet using the [MNIST dataset](http://yann.lecun.com/exdb/ network.set_param_ps() ``` -4. [optional configuration] For a large shape `embedding_table`, because the device can not store a full amount of `embedding_table`. You can configure the `vocab_cache_size` of [EmbeddingLookup operator](https://www.mindspore.cn/docs/en/r1.7/api_python/nn/mindspore.nn.EmbeddingLookup.html) to enable the cache function of `EmbeddingLookup` in the Parameter Server training mode. The `vocab_cache_size` of `embedding_table` is trained on device, and a full amount of `embedding_table` is stored in the Server. The `embedding_table` of the next batch is swapped to the cache in advance, and the expired `embedding_table` is put back to the Server when the cache cannot be placed, to achieve the purpose of improving the training performance. Each Server could save a checkpoint containing the trained `embedding_table` after the training. Detailed network training script can be referred to . +4. [optional configuration] For a large shape `embedding_table`, because the device can not store a full amount of `embedding_table`. You can configure the `vocab_cache_size` of [EmbeddingLookup operator](https://www.mindspore.cn/docs/en/r1.7/api_python/nn/mindspore.nn.EmbeddingLookup.html) to enable the cache function of `EmbeddingLookup` in the Parameter Server training mode. The `vocab_cache_size` of `embedding_table` is trained on device, and a full amount of `embedding_table` is stored in the Server. The `embedding_table` of the next batch is swapped to the cache in advance, and the expired `embedding_table` is put back to the Server when the cache cannot be placed, to achieve the purpose of improving the training performance. Each Server could save a checkpoint containing the trained `embedding_table` after the training. Detailed network training script can be referred to . ```python context.set_auto_parallel_context(full_batch=True, diff --git a/docs/mindspore/source_en/design/recompute.md b/docs/mindspore/source_en/design/recompute.md index 574352839a584b9abab978f2b9b6d158d8ba7020..21158f7ca282df155e4d806823f9276a563b57f4 100644 --- a/docs/mindspore/source_en/design/recompute.md +++ b/docs/mindspore/source_en/design/recompute.md @@ -10,7 +10,7 @@ In order to solve this problem, Mindspore provides the recomputation function. I ## Preliminaries -1. Prepare the model. The ResNet-50 code can be found at: , in which `train.py` is the main function for training, `src/` directory contains the model definition and configuration files of ResNet-50, `script/` directory contains the training and evaluation scripts. +1. Prepare the model. The ResNet-50 code can be found at: , in which `train.py` is the main function for training, `src/` directory contains the model definition and configuration files of ResNet-50, `script/` directory contains the training and evaluation scripts. 2. Prepare the dataset. This example uses the `CIFAR-10` dataset. For details about how to download and load the dataset, visit . diff --git a/docs/mindspore/source_en/faq/data_processing.md b/docs/mindspore/source_en/faq/data_processing.md index 8e0e03184db08b2dbe9808476e60ab509f4463ab..d60c421f59519e2b7237a60fbc9dafd3c47aaa8a 100644 --- a/docs/mindspore/source_en/faq/data_processing.md +++ b/docs/mindspore/source_en/faq/data_processing.md @@ -148,13 +148,13 @@ When `dataset_sink_mode` is set to `False`, data processing and network computin **Q: Can MindSpore train image data of different sizes by batch?** -A: You can refer to the usage of YOLOv3 which contains the resizing of different images. For details about the script, see [yolo_dataset](https://gitee.com/mindspore/models/blob/master/official/cv/yolov3_darknet53/src/yolo_dataset.py). +A: You can refer to the usage of YOLOv3 which contains the resizing of different images. For details about the script, see [yolo_dataset](https://gitee.com/mindspore/models/blob/r1.7/official/cv/yolov3_darknet53/src/yolo_dataset.py).
**Q: Must data be converted into MindRecords when MindSpore is used for segmentation training?** -A: [build_seg_data.py](https://gitee.com/mindspore/models/blob/master/official/cv/deeplabv3/src/data/build_seg_data.py) is the script of MindRecords generated by the dataset. You can directly use or adapt it to your dataset. Alternatively, you can use `GeneratorDataset` to customize the dataset loading if you want to implement the dataset reading by yourself. +A: [build_seg_data.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/deeplabv3/src/data/build_seg_data.py) is the script of MindRecords generated by the dataset. You can directly use or adapt it to your dataset. Alternatively, you can use `GeneratorDataset` to customize the dataset loading if you want to implement the dataset reading by yourself. [GenratorDataset example](https://www.mindspore.cn/tutorials/en/r1.7/advanced/dataset.html) diff --git a/docs/mindspore/source_en/faq/feature_advice.md b/docs/mindspore/source_en/faq/feature_advice.md index d00ac65ce0226f6598a3f6b8042d015caef0ff20..61abbac6b541ada7ccb0025adec5c0359035f9aa 100644 --- a/docs/mindspore/source_en/faq/feature_advice.md +++ b/docs/mindspore/source_en/faq/feature_advice.md @@ -102,7 +102,7 @@ A: MindSpore supports Python native expression and `import mindspore` related pa **Q: Does MindSpore support truncated gradient?** -A: Yes. For details, see [Definition and Usage of Truncated Gradient](https://gitee.com/mindspore/models/blob/master/official/nlp/transformer/src/transformer_for_train.py#L35). +A: Yes. For details, see [Definition and Usage of Truncated Gradient](https://gitee.com/mindspore/models/blob/r1.7/official/nlp/transformer/src/transformer_for_train.py#L35).
@@ -148,7 +148,7 @@ A: PyNative mode is compatible with transfer learning.
-**Q: What is the difference between [MindSpore ModelZoo](https://gitee.com/mindspore/models/blob/master/README.md#) and [Ascend ModelZoo](https://www.hiascend.com/software/modelzoo)?** +**Q: What is the difference between [MindSpore ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README.md#) and [Ascend ModelZoo](https://www.hiascend.com/software/modelzoo)?** A: `MindSpore ModelZoo` contains models mainly implemented by MindSpore. But these models support different devices including Ascend, GPU, CPU and Mobile. `Ascend ModelZoo` contains models only running on Ascend which are implemented by different ML platform including MindSpore, PyTorch, TensorFlow and Caffe. You can refer to the corresponding [gitee repository](https://gitee.com/ascend/modelzoo). diff --git a/docs/mindspore/source_en/faq/implement_problem.md b/docs/mindspore/source_en/faq/implement_problem.md index 60f9bf2e6fbc5ee6a7d801370ca7361e6939a352..8171aabf2cd3ca7d9a8410696b015a42223d6f10 100644 --- a/docs/mindspore/source_en/faq/implement_problem.md +++ b/docs/mindspore/source_en/faq/implement_problem.md @@ -4,7 +4,7 @@ **Q: How do I use MindSpore to implement multi-scale training?** -A: During multi-scale training, when different `shape` are used to call `Cell` objects, different graphs are automatically built and called based on different `shape`, to implement the multi-scale training. Note that multi-scale training supports only the non-data sink mode and does not support the data offloading mode. For details, see the multi-scale training implement of [yolov3](https://gitee.com/mindspore/models/tree/master/official/cv/yolov3_darknet53). +A: During multi-scale training, when different `shape` are used to call `Cell` objects, different graphs are automatically built and called based on different `shape`, to implement the multi-scale training. Note that multi-scale training supports only the non-data sink mode and does not support the data offloading mode. For details, see the multi-scale training implement of [yolov3](https://gitee.com/mindspore/models/tree/r1.7/official/cv/yolov3_darknet53).
@@ -94,7 +94,7 @@ def count_params(net): return total_params ``` -[Script Link](https://gitee.com/mindspore/models/blob/master/research/cv/tinynet/src/utils.py). +[Script Link](https://gitee.com/mindspore/models/blob/r1.7/research/cv/tinynet/src/utils.py).
@@ -283,7 +283,7 @@ A: First, enter the `PTH` file of PyTorch. Taking `ResNet-18` as an example, the **Q: What are the available recommendation or text generation networks or models provided by MindSpore?** -A: Currently, recommendation models such as Wide & Deep, DeepFM, and NCF are under development. In the natural language processing (NLP) field, Bert\_NEZHA is available and models such as MASS are under development. You can rebuild the network into a text generation network based on the scenario requirements. Please stay tuned for updates on the [MindSpore ModelZoo](https://gitee.com/mindspore/models/blob/master/README.md#). +A: Currently, recommendation models such as Wide & Deep, DeepFM, and NCF are under development. In the natural language processing (NLP) field, Bert\_NEZHA is available and models such as MASS are under development. You can rebuild the network into a text generation network based on the scenario requirements. Please stay tuned for updates on the [MindSpore ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README.md#).
diff --git a/docs/mindspore/source_en/migration_guide/performance_optimization.md b/docs/mindspore/source_en/migration_guide/performance_optimization.md index 1e1b03d97721bdde8a9ad78696a6f14eddf031ec..f50a6b24a0cb36c5e6e195b0401b0f2a5e52d654 100644 --- a/docs/mindspore/source_en/migration_guide/performance_optimization.md +++ b/docs/mindspore/source_en/migration_guide/performance_optimization.md @@ -20,7 +20,7 @@ This section will introduce the common use of MindSpore Profiler through three t ### Case 1: Long Step Interval -We run ResNet50 training script in MindSpore [ModelZoo](https://gitee.com/mindspore/models/tree/master ) with batch size set to 32, and we find that each step cost almost 90ms, with a poor performance. +We run ResNet50 training script in MindSpore [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7 ) with batch size set to 32, and we find that each step cost almost 90ms, with a poor performance. As we observed on the MindInsight UI page, the step interval in the step trace is too long, which may indicate that data is the performance bottleneck. ![long_step_interval](images/profiler_case1_long_step_interval.png) @@ -81,7 +81,7 @@ By observing the step trace on the MindInsight performance analysis page, you ca ### Case 2: Long Forward Propagation Interval -We run VGG16 inference script in MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/master/README.md#) , and each step cost almost 113.79ms, with a poor performance. +We run VGG16 inference script in MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README.md#) , and each step cost almost 113.79ms, with a poor performance. As we observed on the MindInsight UI page, the forward propagation in the step trace is too long, which may indicate that operators performance can be optimized. In a single card training or inference process, the forward time consumption is usually considered whether there is a operator that can be optimized for the time consumption. @@ -114,7 +114,7 @@ After the float16 format is set, the inference script is run. From the MindInsig ### Case 3: Optimize The Step Tail -We run ResNet50 training script with 8 processes in MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/master/README.md#) , set batch size to 32, and each step cost about 23.6ms. We still want to improve each step time consumption. +We run ResNet50 training script with 8 processes in MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README.md#) , set batch size to 32, and each step cost about 23.6ms. We still want to improve each step time consumption. As we observed the step trace on the MindInsight UI page, step interval and FP/BP interval can not be improved more, so we try to optimize step tail. diff --git a/docs/mindspore/source_en/migration_guide/preparation.md b/docs/mindspore/source_en/migration_guide/preparation.md index 4be6573b0fcbc84d54c5bc6e52d09a46d0a08c81..11942a27321c9f4d29d760aa24ad28774297e74c 100644 --- a/docs/mindspore/source_en/migration_guide/preparation.md +++ b/docs/mindspore/source_en/migration_guide/preparation.md @@ -44,7 +44,7 @@ Users can read [MindSpore Tutorials](https://www.mindspore.cn/tutorials/experts/ ### ModelZoo and Hub -[ModelZoo](https://gitee.com/mindspore/models/blob/master/README.md#) is a model market of MindSpore and community, which provides deeply-optimized models to developers. In order that the users of MindSpore will have individual development conveniently based on models in ModelZoo. Currently, there are major models in several fields, like computer vision, natural language processing, audio and recommender systems. +[ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README.md#) is a model market of MindSpore and community, which provides deeply-optimized models to developers. In order that the users of MindSpore will have individual development conveniently based on models in ModelZoo. Currently, there are major models in several fields, like computer vision, natural language processing, audio and recommender systems. [mindspore Hub](https://www.mindspore.cn/resources/hub/en) is a platform to save pretrained model of official MindSpore or third party developers. It provides some simple and useful APIs for developers to load and finetune models, so that users can infer or tune models based on pretrained models and deploy models to their applications. Users is able to follow some steps to [publish model](https://www.mindspore.cn/hub/docs/en/r1.6/publish_model.html) to MindSpore Hub,for other developers to download and use. diff --git a/docs/mindspore/source_en/migration_guide/sample_code.md b/docs/mindspore/source_en/migration_guide/sample_code.md index 724255713d632e754c223817d513933cf0565cc1..16cd23f9d2df39c336dea55d71588bc322c03325 100644 --- a/docs/mindspore/source_en/migration_guide/sample_code.md +++ b/docs/mindspore/source_en/migration_guide/sample_code.md @@ -665,7 +665,7 @@ if __name__ == '__main__': model.train(config.epoch_size, dataset, callbacks=cb, sink_size=step_size, dataset_sink_mode=False) ``` -Note: For codes in other files in the directory, refer to MindSpore ModelZoo's [ResNet50 implementation](https://gitee.com/mindspore/models/tree/master/official/cv/resnet)(this script incorporates other ResNet family networks and ResNet-SE networks, and the specific implementation may differ from the benchmark script). +Note: For codes in other files in the directory, refer to MindSpore ModelZoo's [ResNet50 implementation](https://gitee.com/mindspore/models/tree/r1.7/official/cv/resnet)(this script incorporates other ResNet family networks and ResNet-SE networks, and the specific implementation may differ from the benchmark script). ### Distributed Training @@ -835,7 +835,7 @@ For MindData performance issues, refer to MindData in MindInsight Component's [D When distributed training is performed, after the forward propagation and gradient computation are completed during a Step, each machine starts to synchronize the AllReduce gradient, and the AllReduce synchronization time is mainly affected by the number of weights and machines. For more complex, larger machine-sized networks, the AllReduce gradient update time is longer, at which point we can perform AllReduce tangent to optimize this part of the time. Normally, AllReduce gradient synchronization waits until all the inverse operators are finished, i.e., all the gradients of all weights are computed before synchronizing the gradients of all machines at once, but with AllReduce tangent, we can synchronize the gradients of some weights as soon as they are computed, so that the gradient synchronization and the gradient computation of the remaining operators can be performed in parallel, hiding this part of the AllReduce gradient synchronization time. The tangent strategy is usually a manual attempt to find an optimal solution (supporting slicing greater than two segments). -As an example, [ResNet50 network](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/train.py) has 160 weights and [85, 160] means that the gradient synchronization is performed immediately after the gradient is calculated for the 0th to 85th weights, and the gradient synchronization is performed after the gradient is calculated for the 86th to 160th weights. Here the two segments is sliced, so two gradient synchronizations are required. The code implementation is as follows: +As an example, [ResNet50 network](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/train.py) has 160 weights and [85, 160] means that the gradient synchronization is performed immediately after the gradient is calculated for the 0th to 85th weights, and the gradient synchronization is performed after the gradient is calculated for the 86th to 160th weights. Here the two segments is sliced, so two gradient synchronizations are required. The code implementation is as follows: ```python device_id = int(os.getenv('DEVICE_ID', '0')) diff --git a/docs/mindspore/source_en/note/benchmark.md b/docs/mindspore/source_en/note/benchmark.md index c329d8b55e77e7a243646ec66460afd5c0be2684..d9bd77700a930fafa379aff5d020e0d061880be5 100644 --- a/docs/mindspore/source_en/note/benchmark.md +++ b/docs/mindspore/source_en/note/benchmark.md @@ -3,7 +3,7 @@ This document describes the MindSpore benchmarks. -For details about the MindSpore networks, see [ModelZoo](https://gitee.com/mindspore/models/blob/master/README.md#). +For details about the MindSpore networks, see [ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README.md#). ## Training Performance diff --git a/docs/mindspore/source_en/note/network_list_ms.md b/docs/mindspore/source_en/note/network_list_ms.md index fd234331289cc99ba2838862ce2995207d63ed3d..a9d91cb30cc68fba8e8bc590c98d92d22da27649 100644 --- a/docs/mindspore/source_en/note/network_list_ms.md +++ b/docs/mindspore/source_en/note/network_list_ms.md @@ -2,6 +2,6 @@ -Please obtain the [Network List](https://gitee.com/mindspore/models/blob/master/README.md#table-of-contents) from ModelZoo. +Please obtain the [Network List](https://gitee.com/mindspore/models/blob/r1.7/README.md#table-of-contents) from ModelZoo. You can also use [MindWizard Tool](https://gitee.com/mindspore/mindinsight/tree/r1.7/mindinsight/wizard/) to quickly generate classic network scripts. diff --git a/docs/mindspore/source_zh_cn/design/comm_fusion.md b/docs/mindspore/source_zh_cn/design/comm_fusion.md index 733b92bad78439b52c272a5ae2bef0d0d7738e89..c93696a065ad252f822360eecee7a588c8702463 100644 --- a/docs/mindspore/source_zh_cn/design/comm_fusion.md +++ b/docs/mindspore/source_zh_cn/design/comm_fusion.md @@ -169,7 +169,7 @@ The parameter layer3.output_mapping.bias's fusion id is 2 Ascend分布式相关的环境变量有: -- RANK_TABLE_FILE：组网信息文件的路径。rank_table_file文件可以使用models代码仓中的hccl_tools.py生成，可以从[此处](https://gitee.com/mindspore/models/tree/master/utils/hccl_tools)获取。 +- RANK_TABLE_FILE：组网信息文件的路径。rank_table_file文件可以使用models代码仓中的hccl_tools.py生成，可以从[此处](https://gitee.com/mindspore/models/tree/r1.7/utils/hccl_tools)获取。 - DEVICE_ID：当前卡在机器上的实际序号。 - RANK_ID：当前卡的逻辑序号。 diff --git a/docs/mindspore/source_zh_cn/design/cv_resnet50_second_order_optimizer.md b/docs/mindspore/source_zh_cn/design/cv_resnet50_second_order_optimizer.md index cb2a37bcdd857bc2249696f092c18286ac4090b3..c50c0e9f200f043f9599905d8b06195c33c59dcf 100644 --- a/docs/mindspore/source_zh_cn/design/cv_resnet50_second_order_optimizer.md +++ b/docs/mindspore/source_zh_cn/design/cv_resnet50_second_order_optimizer.md @@ -10,7 +10,7 @@ MindSpore开发团队在现有的自然梯度算法的基础上，对FIM矩阵 本篇教程将主要介绍如何在Ascend 910 以及GPU上，使用MindSpore提供的二阶优化器THOR训练ResNet50-v1.5网络和ImageNet数据集。 > 你可以在这里下载完整的示例代码： - 。 + 。 示例代码目录结构 @@ -173,7 +173,7 @@ def create_dataset2(dataset_path, do_train, repeat_num=1, batch_size=32, target= ## 定义网络 -本示例中使用的网络模型为ResNet50-v1.5，定义[ResNet50网络](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/src/resnet.py)。 +本示例中使用的网络模型为ResNet50-v1.5，定义[ResNet50网络](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/src/resnet.py)。 网络构建完成以后，在`__main__`函数中调用定义好的ResNet50： @@ -338,7 +338,7 @@ bash run_distribute_train.sh [CONFIG_PATH] 脚本需要传入变量`RANK_TABLE_FILE`，`DATASET_PATH`和`CONFIG_PATH`，其中： -- `RANK_TABLE_FILE`：组网信息文件的路径。(rank table文件的生成，参考[HCCL_TOOL](https://gitee.com/mindspore/models/tree/master/utils/hccl_tools)) +- `RANK_TABLE_FILE`：组网信息文件的路径。(rank table文件的生成，参考[HCCL_TOOL](https://gitee.com/mindspore/models/tree/r1.7/utils/hccl_tools)) - `DATASET_PATH`：训练数据集路径。 - `CONFIG_PATH`：配置文件路径。 diff --git a/docs/mindspore/source_zh_cn/design/heterogeneous_training.ipynb b/docs/mindspore/source_zh_cn/design/heterogeneous_training.ipynb index f81a2d28fe2a4a1425359f790e2e6f8310b863e3..cb10052a4ebf9af9a5cd4c3b5eb938fbb2f3a399 100644 --- a/docs/mindspore/source_zh_cn/design/heterogeneous_training.ipynb +++ b/docs/mindspore/source_zh_cn/design/heterogeneous_training.ipynb @@ -183,7 +183,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "步骤4、5也可以直接融合到优化器算子中做进一步优化，完整的优化器异构训练流程可以参考： \n", + "步骤4、5也可以直接融合到优化器算子中做进一步优化，完整的优化器异构训练流程可以参考： \n", "\n", "## Embedding异构\n", "\n", @@ -291,7 +291,7 @@ "source": [ "当前nn目录下的EmbeddingLookup、FTRL、LazyAdam等算子已经封装好异构接口，用户只需设置target属性为CPU或DEVICE即可切换执行后端。\n", "\n", - "整体调用流程可以参考：\n", + "整体调用流程可以参考：\n", "\n", "## PS异构\n", "\n", @@ -301,7 +301,7 @@ "\n", "Parameter Server封装异构流程，用户只需配置参数使用PS即可，具体配置流程请参考[Parameter Server训练流程](https://www.mindspore.cn/docs/zh-CN/r1.7/design/parameter_server_training.html)。\n", "\n", - "此外，wide&deep网络中也有使用PS的流程，可参考：\n", + "此外，wide&deep网络中也有使用PS的流程，可参考：\n", "\n", "## 约束\n", "\n", diff --git a/docs/mindspore/source_zh_cn/design/host_device_training.md b/docs/mindspore/source_zh_cn/design/host_device_training.md index 9665fce79d011a8e0b952d7cfec3c3dd3eaa3047..a0779c8626892c26da533daedcafdc7e1aa57926 100644 --- a/docs/mindspore/source_zh_cn/design/host_device_training.md +++ b/docs/mindspore/source_zh_cn/design/host_device_training.md @@ -6,7 +6,7 @@ 在深度学习中，工作人员时常会遇到超大模型的训练问题，即模型参数所占内存超过了设备内存上限。为高效地训练超大模型，一种方案便是分布式并行训练，也就是将工作交由同构的多个加速器（如Ascend 910 AI处理器，GPU等）共同完成。但是这种方式在面对几百GB甚至几TB级别的模型时，所需的加速器过多。而当从业者实际难以获取大规模集群时，这种方式难以应用。另一种可行的方案是使用主机端（Host）和加速器（Device）的混合训练模式。此方案同时发挥了主机端内存大和加速器端计算快的优势，是一种解决超大模型训练较有效的方式。 -在MindSpore中，用户可以将待训练的参数放在主机，同时将必要算子的执行位置配置为主机，其余算子的执行位置配置为加速器，从而方便地实现混合训练。此教程以推荐模型[Wide&Deep](https://gitee.com/mindspore/models/tree/master/official/recommend/wide_and_deep)为例，讲解MindSpore在主机和Ascend 910 AI处理器的混合训练。 +在MindSpore中，用户可以将待训练的参数放在主机，同时将必要算子的执行位置配置为主机，其余算子的执行位置配置为加速器，从而方便地实现混合训练。此教程以推荐模型[Wide&Deep](https://gitee.com/mindspore/models/tree/r1.7/official/recommend/wide_and_deep)为例，讲解MindSpore在主机和Ascend 910 AI处理器的混合训练。 ## 基本原理 @@ -26,11 +26,11 @@ ### 样例代码说明 -1. 准备模型代码。Wide&Deep的代码可参见：，其中，`train_and_eval_auto_parallel.py`脚本定义了模型训练的主流程，`src/`目录中包含Wide&Deep模型的定义、数据处理和配置信息等，`script/`目录中包含不同配置下的训练脚本。 +1. 准备模型代码。Wide&Deep的代码可参见：，其中，`train_and_eval_auto_parallel.py`脚本定义了模型训练的主流程，`src/`目录中包含Wide&Deep模型的定义、数据处理和配置信息等，`script/`目录中包含不同配置下的训练脚本。 2. 准备数据集。请参考[1]中的论文所提供的链接下载数据集，并利用脚本`src/preprocess_data.py`将数据集转换为MindRecord格式。 -3. 配置处理器信息。在裸机环境（即本地有Ascend 910 AI 处理器）进行分布式训练时，需要配置加速器信息文件。此样例只使用一个加速器，故只需配置包含0号卡的`rank_table_1p_0.json`文件。MindSpore提供了生成该配置文件的自动化生成脚本及相关说明，可参考[HCCL_TOOL](https://gitee.com/mindspore/models/tree/master/utils/hccl_tools)。 +3. 配置处理器信息。在裸机环境（即本地有Ascend 910 AI 处理器）进行分布式训练时，需要配置加速器信息文件。此样例只使用一个加速器，故只需配置包含0号卡的`rank_table_1p_0.json`文件。MindSpore提供了生成该配置文件的自动化生成脚本及相关说明，可参考[HCCL_TOOL](https://gitee.com/mindspore/models/tree/r1.7/utils/hccl_tools)。 ### 配置混合执行 diff --git a/docs/mindspore/source_zh_cn/design/optimizer_parallel.md b/docs/mindspore/source_zh_cn/design/optimizer_parallel.md index e403eca595305e19faa676defcab9d0f2c3087c2..6af8a7121f11e1ac51f71b1c009a448a169e379b 100644 --- a/docs/mindspore/source_zh_cn/design/optimizer_parallel.md +++ b/docs/mindspore/source_zh_cn/design/optimizer_parallel.md @@ -184,7 +184,7 @@ The parameter layer3.output_mapping.bias's fusion id is 2 Ascend分布式相关的环境变量有: -- RANK_TABLE_FILE：组网信息文件的路径。rank_table_file文件可以使用models代码仓中的hccl_tools.py生成，可以从[此处](https://gitee.com/mindspore/models/tree/master/utils/hccl_tools)获取。 +- RANK_TABLE_FILE：组网信息文件的路径。rank_table_file文件可以使用models代码仓中的hccl_tools.py生成，可以从[此处](https://gitee.com/mindspore/models/tree/r1.7/utils/hccl_tools)获取。 - DEVICE_ID：当前卡在机器上的实际序号。 - RANK_ID：当前卡的逻辑序号。 diff --git a/docs/mindspore/source_zh_cn/design/parameter_server_training.md b/docs/mindspore/source_zh_cn/design/parameter_server_training.md index 8d36595c73445ce4c56f5d1a8be00518197c82da..08300658e4041d4218fb962d6b19936d10b296d3 100644 --- a/docs/mindspore/source_zh_cn/design/parameter_server_training.md +++ b/docs/mindspore/source_zh_cn/design/parameter_server_training.md @@ -24,7 +24,7 @@ MindSpore的参数服务器采用了自研的通信框架作为基础架构， ### 训练脚本准备 -参考，使用[MNIST数据集](http://yann.lecun.com/exdb/mnist/)，了解如何训练一个LeNet网络。 +参考，使用[MNIST数据集](http://yann.lecun.com/exdb/mnist/)，了解如何训练一个LeNet网络。 ### 参数设置 @@ -41,7 +41,7 @@ MindSpore的参数服务器采用了自研的通信框架作为基础架构， - 被设置为通过Parameter Server更新的单个权重大小不得超过INT_MAX(2^31 - 1)字节。 - 接口`set_param_ps`可接收一个`bool`型参数：`init_in_server`，表示该训练参数是否在Server端初始化，`init_in_server`默认值为`False`，表示在Worker上初始化该训练参数；当前仅支持`EmbeddingLookup`算子的训练参数`embedding_table`在Server端初始化，以解决超大shape的`embedding_table`在Worker上初始化导致内存不足的问题，该算子的`target`属性需要设置为'CPU'。在Server端初始化的训练参数将不再同步到Worker上，如果涉及到多Server训练并保存CheckPoint，则训练结束后每个Server均会保存一个CheckPoint。 -3. 在[LeNet原训练脚本](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/train.py)基础上，设置该模型所有权重由Parameter Server训练： +3. 在[LeNet原训练脚本](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/train.py)基础上，设置该模型所有权重由Parameter Server训练： ```python context.set_ps_context(enable_ps=True) @@ -49,7 +49,7 @@ MindSpore的参数服务器采用了自研的通信框架作为基础架构， network.set_param_ps() ``` -4. [可选配置]针对超大shape的`embedding_table`，由于设备上存放不下全量的`embedding_table`，可以配置[EmbeddingLookup算子](https://www.mindspore.cn/docs/zh-CN/r1.7/api_python/nn/mindspore.nn.EmbeddingLookup.html)的`vocab_cache_size`，用于开启Parameter Server训练模式下`EmbeddingLookup`的cache功能，该功能使用`vocab_cache_size`大小的`embedding_table`在设备上训练，全量`embedding_table`存储在Server，将下批次训练用到的`embedding_table`提前换入到cache上，当cache放不下时则将过期的`embedding_table`放回到Server，以达到提升训练性能的目的；训练结束后，可在Server上导出CheckPoint，保存训练后的全量`embedding_table`。详细网络训练脚本参考。 +4. [可选配置]针对超大shape的`embedding_table`，由于设备上存放不下全量的`embedding_table`，可以配置[EmbeddingLookup算子](https://www.mindspore.cn/docs/zh-CN/r1.7/api_python/nn/mindspore.nn.EmbeddingLookup.html)的`vocab_cache_size`，用于开启Parameter Server训练模式下`EmbeddingLookup`的cache功能，该功能使用`vocab_cache_size`大小的`embedding_table`在设备上训练，全量`embedding_table`存储在Server，将下批次训练用到的`embedding_table`提前换入到cache上，当cache放不下时则将过期的`embedding_table`放回到Server，以达到提升训练性能的目的；训练结束后，可在Server上导出CheckPoint，保存训练后的全量`embedding_table`。详细网络训练脚本参考。 ```python context.set_auto_parallel_context(full_batch=True, parallel_mode=ParallelMode.AUTO_PARALLEL) diff --git a/docs/mindspore/source_zh_cn/design/pynative_shard_function_parallel.md b/docs/mindspore/source_zh_cn/design/pynative_shard_function_parallel.md index a529f81fb10e2b771f1bb89e8d44d1b05c775037..d8850fc3d1fd9944845719526d42e5a035745e2e 100644 --- a/docs/mindspore/source_zh_cn/design/pynative_shard_function_parallel.md +++ b/docs/mindspore/source_zh_cn/design/pynative_shard_function_parallel.md @@ -173,7 +173,7 @@ class Net(nn.Cell): Ascend分布式相关的环境变量有: -- RANK_TABLE_FILE：组网信息文件的路径。rank_table_file文件可以使用models代码仓中的hccl_tools.py生成，可以从[此处](https://gitee.com/mindspore/models/tree/master/utils/hccl_tools)获取。 +- RANK_TABLE_FILE：组网信息文件的路径。rank_table_file文件可以使用models代码仓中的hccl_tools.py生成，可以从[此处](https://gitee.com/mindspore/models/tree/r1.7/utils/hccl_tools)获取。 - DEVICE_ID：当前卡在机器上的实际序号。 - RANK_ID：当前卡的逻辑序号。 diff --git a/docs/mindspore/source_zh_cn/design/recompute.md b/docs/mindspore/source_zh_cn/design/recompute.md index 0b9dad885be69fc1706cce22bcfca52e4433dd5f..347677fd96e362ef80bb030d757264980c6ecf6c 100644 --- a/docs/mindspore/source_zh_cn/design/recompute.md +++ b/docs/mindspore/source_zh_cn/design/recompute.md @@ -32,7 +32,7 @@ MindSpore根据正向图计算流程来自动推导出反向图，正向图和 ### 样例代码说明 -1. 准备模型代码。ResNet-50模型的代码可参见：，其中，`train.py`为训练的主函数所在，`src/`目录中包含ResNet-50模型的定义和配置信息等，`script/`目录中包含一些训练和推理脚本。 +1. 准备模型代码。ResNet-50模型的代码可参见：，其中，`train.py`为训练的主函数所在，`src/`目录中包含ResNet-50模型的定义和配置信息等，`script/`目录中包含一些训练和推理脚本。 2. 准备数据集。本样例采用`CIFAR-10`数据集，数据集的下载和加载方式可参考：。 ### 配置重计算 diff --git a/docs/mindspore/source_zh_cn/faq/data_processing.md b/docs/mindspore/source_zh_cn/faq/data_processing.md index c064b44bd16c685dd49d3af3f35ab013dc888f7e..bd487575c093cb1313850e17eb5f3fa4717cdecb 100644 --- a/docs/mindspore/source_zh_cn/faq/data_processing.md +++ b/docs/mindspore/source_zh_cn/faq/data_processing.md @@ -148,13 +148,13 @@ A: 当`dataset_sink_mode=True`时，数据处理会和网络计算构成Pipeline **Q: MindSpore能否支持按批次对不同尺寸的图片数据进行训练？** -A: 你可以参考yolov3对于此场景的使用，里面有对于图像的不同缩放,脚本见[yolo_dataset](https://gitee.com/mindspore/models/blob/master/official/cv/yolov3_darknet53/src/yolo_dataset.py)。 +A: 你可以参考yolov3对于此场景的使用，里面有对于图像的不同缩放,脚本见[yolo_dataset](https://gitee.com/mindspore/models/blob/r1.7/official/cv/yolov3_darknet53/src/yolo_dataset.py)。
**Q: 使用MindSpore做分割训练，必须将数据转为MindRecord吗？** -A: [build_seg_data.py](https://gitee.com/mindspore/models/blob/master/official/cv/deeplabv3/src/data/build_seg_data.py)是将数据集生成MindRecord的脚本，可以直接使用/适配下你的数据集。或者如果你想尝试自己实现数据集的读取，可以使用`GeneratorDataset`自定义数据集加载。 +A: [build_seg_data.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/deeplabv3/src/data/build_seg_data.py)是将数据集生成MindRecord的脚本，可以直接使用/适配下你的数据集。或者如果你想尝试自己实现数据集的读取，可以使用`GeneratorDataset`自定义数据集加载。 [GenratorDataset 示例](https://www.mindspore.cn/tutorials/zh-CN/r1.7/advanced/dataset/custom.html) diff --git a/docs/mindspore/source_zh_cn/faq/feature_advice.md b/docs/mindspore/source_zh_cn/faq/feature_advice.md index dc7ca6a3ba31ab6a0052bd646ae08e4c9a8dacae..d493fe2a42bc07291ec0a906086a6039ba0808bb 100644 --- a/docs/mindspore/source_zh_cn/faq/feature_advice.md +++ b/docs/mindspore/source_zh_cn/faq/feature_advice.md @@ -102,7 +102,7 @@ A: MindSpore支持Python原生表达，`import mindspore`相关包即可使用 **Q: 请问MindSpore支持梯度截断吗？** -A: 支持，可以参考[梯度截断的定义和使用](https://gitee.com/mindspore/models/blob/master/official/nlp/transformer/src/transformer_for_train.py#L35)。 +A: 支持，可以参考[梯度截断的定义和使用](https://gitee.com/mindspore/models/blob/r1.7/official/nlp/transformer/src/transformer_for_train.py#L35)。
@@ -148,7 +148,7 @@ A: PyNative模式是兼容迁移学习的。
-**Q: MindSpore仓库中的[ModelZoo](https://gitee.com/mindspore/models/blob/master/README_CN.md#)和昇腾官网的[ModelZoo](https://www.hiascend.com/software/modelzoo)有什么关系？** +**Q: MindSpore仓库中的[ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README_CN.md#)和昇腾官网的[ModelZoo](https://www.hiascend.com/software/modelzoo)有什么关系？** A: MindSpore的ModelZoo主要提供MindSpore框架实现的模型，同时包括了Ascend/GPU/CPU/Mobile多种设备的支持。昇腾的ModelZoo主要提供运行于Ascend加速芯片上的模型，包括了MindSpore/PyTorch/TensorFlow/Caffe等多种框架的支持。可以参考对应的[Gitee仓库](https://gitee.com/ascend/modelzoo) diff --git a/docs/mindspore/source_zh_cn/faq/implement_problem.md b/docs/mindspore/source_zh_cn/faq/implement_problem.md index cb8b970fb9e6680dca75f58fd54aa34fa3fc13cc..3be6abddff688de456b625afb9f242ed53507509 100644 --- a/docs/mindspore/source_zh_cn/faq/implement_problem.md +++ b/docs/mindspore/source_zh_cn/faq/implement_problem.md @@ -4,7 +4,7 @@ **Q: 请问使用MindSpore如何实现多尺度训练？** -A: 在多尺度训练过程中，使用不同`shape`调用`Cell`对象的时候，会自动根据不同`shape`编译并调用不同的图，从而实现多尺度的训练。要注意多尺度训练只支持非数据下沉模式，不能支持数据下沉的训练方式。可以参考[yolov3](https://gitee.com/mindspore/models/tree/master/official/cv/yolov3_darknet53)的多尺度训练实现。 +A: 在多尺度训练过程中，使用不同`shape`调用`Cell`对象的时候，会自动根据不同`shape`编译并调用不同的图，从而实现多尺度的训练。要注意多尺度训练只支持非数据下沉模式，不能支持数据下沉的训练方式。可以参考[yolov3](https://gitee.com/mindspore/models/tree/r1.7/official/cv/yolov3_darknet53)的多尺度训练实现。
@@ -94,7 +94,7 @@ def count_params(net): return total_params ``` -具体[脚本链接](https://gitee.com/mindspore/models/blob/master/research/cv/tinynet/src/utils.py)。 +具体[脚本链接](https://gitee.com/mindspore/models/blob/r1.7/research/cv/tinynet/src/utils.py)。
@@ -267,7 +267,7 @@ A: 首先输入PyTorch的`pth`文件，以`ResNet-18`为例，MindSpore的网络 **Q: MindSpore有哪些现成的推荐类或生成类网络或模型可用？** -A: 目前正在开发Wide & Deep、DeepFM、NCF等推荐类模型，NLP领域已经支持Bert_NEZHA，正在开发MASS等模型，用户可根据场景需要改造为生成类网络，可以关注[MindSpore ModelZoo](https://gitee.com/mindspore/models/blob/master/README_CN.md#)。 +A: 目前正在开发Wide & Deep、DeepFM、NCF等推荐类模型，NLP领域已经支持Bert_NEZHA，正在开发MASS等模型，用户可根据场景需要改造为生成类网络，可以关注[MindSpore ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README_CN.md#)。
diff --git a/docs/mindspore/source_zh_cn/migration_guide/performance_optimization.md b/docs/mindspore/source_zh_cn/migration_guide/performance_optimization.md index 3bba16ffc49eb742618e3e511945306d9a975fde..01eb26360424df3a1e1326c75a99252ef2edc2dd 100644 --- a/docs/mindspore/source_zh_cn/migration_guide/performance_optimization.md +++ b/docs/mindspore/source_zh_cn/migration_guide/performance_optimization.md @@ -20,7 +20,7 @@ Profiler的功能介绍及使用说明请参见教程： ### 案例一：迭代间隙过长 -在MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/master/README_CN.md#)中运行ResNet50单卡训练脚本，batch size设置为32，发现单step时间约为90ms，性能较差。 +在MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README_CN.md#)中运行ResNet50单卡训练脚本，batch size设置为32，发现单step时间约为90ms，性能较差。 通过MindInsight性能分析页面观察到迭代轨迹中的迭代间隙过长，这通常说明数据是性能瓶颈点。 ![long_step_interval](images/profiler_case1_long_step_interval.png) @@ -78,7 +78,7 @@ data_set = data_set.map(operations=trans, input_columns="image", num_parallel_wo ### 案例二：前向运行时间长 -在MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/master/README_CN.md#)中运行VGG16模型的推理脚本，发现单step时间约为113.79ms，性能较差。 +在MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README_CN.md#)中运行VGG16模型的推理脚本，发现单step时间约为113.79ms，性能较差。 通过MindInsight性能分析页面观察到迭代轨迹中的前向运行时间很长。在单卡训练或推理过程中，前向耗时长通常考虑是否有算子的耗时时长可以优化。 ![long_fp_bp](images/profiler_case2_long_fpbp.png) @@ -110,7 +110,7 @@ network.add_flags_recursive(fp16=True) ### 案例三： 优化迭代拖尾 -在MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/master/README_CN.md#)中运行ResNet50 8卡训练脚本，batch size设置为32，单step时间为23.6ms，期望能继续提高单step时间。 +在MindSpore [ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README_CN.md#)中运行ResNet50 8卡训练脚本，batch size设置为32，单step时间为23.6ms，期望能继续提高单step时间。 通过MindInsight性能分析页面观察迭代轨迹，发现迭代间隙与前反向已经没有多少优化的空间，考虑迭代拖尾是否可以优化。 ![long_tail](images/profiler_case3_long_tail.png) diff --git a/docs/mindspore/source_zh_cn/migration_guide/preparation.ipynb b/docs/mindspore/source_zh_cn/migration_guide/preparation.ipynb index 7fad5f08f3d5b03b5c7450048b8c4692d81d8937..427eaf33b56dc67929317a7afdfd4d62e77fbc35 100644 --- a/docs/mindspore/source_zh_cn/migration_guide/preparation.ipynb +++ b/docs/mindspore/source_zh_cn/migration_guide/preparation.ipynb @@ -69,7 +69,7 @@ "\n", "### ModelZoo和Hub\n", "\n", - "[ModelZoo](https://gitee.com/mindspore/models/blob/master/README_CN.md#)是MindSpore与社区共同提供的深度优化的模型集市，向开发者提供了深度优化的模型，以便于生态中的小伙伴可以方便地基于ModelZoo中的模型进行个性化开发。当前已经覆盖了机器视觉、自然语言处理、语音、推荐系统等多个领域的主流模型。\n", + "[ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README_CN.md#)是MindSpore与社区共同提供的深度优化的模型集市，向开发者提供了深度优化的模型，以便于生态中的小伙伴可以方便地基于ModelZoo中的模型进行个性化开发。当前已经覆盖了机器视觉、自然语言处理、语音、推荐系统等多个领域的主流模型。\n", "\n", "[mindspore Hub](https://www.mindspore.cn/resources/hub)是存放MindSpore官方或者第三方开发者提供的预训练模型的平台。它向应用开发者提供了简单易用的模型加载和微调API，使得用户可以基于预训练模型进行推理或者微调，并部署到自己的应用中。用户也可以将自己训练好的模型按照指定的步骤[发布模型](https://www.mindspore.cn/hub/docs/zh-CN/r1.6/publish_model.html)到MindSpore Hub中，供其他用户下载和使用。\n", "\n", diff --git a/docs/mindspore/source_zh_cn/migration_guide/sample_code.md b/docs/mindspore/source_zh_cn/migration_guide/sample_code.md index 9ed823c63c0dfeeb746da57f76edbaf4b5b1b68c..ae32083bd550fa5d40d1df270d84098320d871b0 100644 --- a/docs/mindspore/source_zh_cn/migration_guide/sample_code.md +++ b/docs/mindspore/source_zh_cn/migration_guide/sample_code.md @@ -662,7 +662,7 @@ if __name__ == '__main__': model.train(config.epoch_size, dataset, callbacks=cb, sink_size=step_size, dataset_sink_mode=False) ``` -注意：关于目录中其他文件的代码，可以参考 MindSpore ModelZoo 的 [ResNet50 实现](https://gitee.com/mindspore/models/tree/master/official/cv/resnet)（该脚本融合了其他 ResNet 系列网络及ResNet-SE 网络，具体实现可能和对标脚本有差异）。 +注意：关于目录中其他文件的代码，可以参考 MindSpore ModelZoo 的 [ResNet50 实现](https://gitee.com/mindspore/models/tree/r1.7/official/cv/resnet)（该脚本融合了其他 ResNet 系列网络及ResNet-SE 网络，具体实现可能和对标脚本有差异）。 ### 分布式训练 @@ -830,7 +830,7 @@ profiler.analyse() 当进行分布式训练时，在一个Step的训练过程中，完成前向传播和梯度计算后，各个机器开始进行AllReduce梯度同步，AllReduce同步时间主要受权重数量、机器数量影响，对于越复杂、机器规模越大的网络，其 AllReduce 梯度更新时间也越久，此时我们可以进行AllReduce 切分来优化这部分耗时。 正常情况下，AllReduce 梯度同步会等所有反向算子执行结束，也就是对所有权重都计算出梯度后再一次性同步所有机器的梯度，而使用AllReduce切分后，我们可以在计算出一部分权重的梯度后，就立刻进行这部分权重的梯度同步，这样梯度同步和剩余算子的梯度计算可以并行执行，也就隐藏了这部分 AllReduce 梯度同步时间。切分策略通常是手动尝试，寻找一个最优的方案（支持切分大于两段）。 -以 [ResNet50网络](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/train.py) 为例，该网络共有 160 个 权重， [85, 160] 表示第 0 至 85个权重计算完梯度后立刻进行梯度同步，第 86 至 160 个 权重计算完后再进行梯度同步，这里共切分两段，因此需要进行两次梯度同步。代码实现如下： +以 [ResNet50网络](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/train.py) 为例，该网络共有 160 个 权重， [85, 160] 表示第 0 至 85个权重计算完梯度后立刻进行梯度同步，第 86 至 160 个 权重计算完后再进行梯度同步，这里共切分两段，因此需要进行两次梯度同步。代码实现如下： ```python device_id = int(os.getenv('DEVICE_ID', '0')) diff --git a/docs/mindspore/source_zh_cn/migration_guide/training_process_comparision.md b/docs/mindspore/source_zh_cn/migration_guide/training_process_comparision.md index 8021b045ff0b8e5016715bdb3b9597273566525d..bcc3794588004432207aa634aed7ae496b7aeb13 100644 --- a/docs/mindspore/source_zh_cn/migration_guide/training_process_comparision.md +++ b/docs/mindspore/source_zh_cn/migration_guide/training_process_comparision.md @@ -46,7 +46,7 @@ MindSpore 的模型训练和推理的总体执行流程，基本与主流的 AI model.train(epoch_size, ds_train, callbacks=[loss_cb, ckpoint_cb ]) ``` - 代码来源： [ModelZoo/LeNet5](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/train.py) + 代码来源： [ModelZoo/LeNet5](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/train.py) - PyTorch diff --git a/docs/mindspore/source_zh_cn/note/benchmark.md b/docs/mindspore/source_zh_cn/note/benchmark.md index f89a0122032dabf2c0e1393263fbae344844e35e..ca240fc5990d12f27489d4e8d619756d340ce918 100644 --- a/docs/mindspore/source_zh_cn/note/benchmark.md +++ b/docs/mindspore/source_zh_cn/note/benchmark.md @@ -2,7 +2,7 @@ -本文介绍MindSpore的基准性能。MindSpore网络定义可参考[ModelZoo](https://gitee.com/mindspore/models/blob/master/README_CN.md#)。 +本文介绍MindSpore的基准性能。MindSpore网络定义可参考[ModelZoo](https://gitee.com/mindspore/models/blob/r1.7/README_CN.md#)。 ## 训练性能 diff --git a/docs/mindspore/source_zh_cn/note/network_list_ms.md b/docs/mindspore/source_zh_cn/note/network_list_ms.md index 6849eeb8d7b4e98f2ddd1f5adec24085d16fde6a..044cfe9d4efcb9df677d51c688e8269330af3d57 100644 --- a/docs/mindspore/source_zh_cn/note/network_list_ms.md +++ b/docs/mindspore/source_zh_cn/note/network_list_ms.md @@ -2,6 +2,6 @@ -请从ModelZoo获取[MindSpore网络支持列表](https://gitee.com/mindspore/models/blob/master/README_CN.md#目录)。 +请从ModelZoo获取[MindSpore网络支持列表](https://gitee.com/mindspore/models/blob/r1.7/README_CN.md#目录)。 你也可以使用 [MindWizard工具](https://gitee.com/mindspore/mindinsight/tree/r1.7/mindinsight/wizard/) 快速生成经典网络脚本。 diff --git a/docs/notebook/test_model_security_membership_inference.ipynb b/docs/notebook/test_model_security_membership_inference.ipynb index 0b463bf19cbc546e0c949d4b849c3a0a229bc62e..8b48700b5f21f2c71f284bba7159e4ea29210948 100644 --- a/docs/notebook/test_model_security_membership_inference.ipynb +++ b/docs/notebook/test_model_security_membership_inference.ipynb @@ -169,7 +169,7 @@ "source": [ "ckpt文件参考MindSpore代码仓库中ModelZoo中VGG16代码将cifar10的代码修改为cifar100的代码进行训练产生\n", "\n", - "训练代码下载：[链接](https://gitee.com/mindspore/models/tree/master/official/cv/vgg16)\n", + "训练代码下载：[链接](https://gitee.com/mindspore/models/tree/r1.7/official/cv/vgg16)\n", "\n", "训练完成的ckpt文件下载地址（百度网盘提取码: jits）： [链接](https://pan.baidu.com/s/10jeLzJ1Sl23gjoc-AZd-Ng)\n", "\n", diff --git a/docs/serving/docs/source_en/serving_distributed_example.md b/docs/serving/docs/source_en/serving_distributed_example.md index 39c9622cb793a0ab81a3cf0b6b185c4c41f0838d..0fb62dad5907022496e4953ab9d3df3792df5fd6 100644 --- a/docs/serving/docs/source_en/serving_distributed_example.md +++ b/docs/serving/docs/source_en/serving_distributed_example.md @@ -44,7 +44,7 @@ export_model - `net.py` contains the definition of MatMul network. - `distributed_inference.py` is used to configure distributed parameters. - `export_model.sh` creates `device` directory on the current host and exports model files corresponding to `device`. -- `rank_table_8pcs.json` is a json file for configuring the multi-cards network. For details, see [rank_table](https://gitee.com/mindspore/models/tree/master/utils/hccl_tools). +- `rank_table_8pcs.json` is a json file for configuring the multi-cards network. For details, see [rank_table](https://gitee.com/mindspore/models/tree/r1.7/utils/hccl_tools). Use [net.py](https://gitee.com/mindspore/serving/blob/r1.7/example/matmul_distributed/export_model/net.py) to construct a network that contains the MatMul and Neg operators. diff --git a/docs/serving/docs/source_en/serving_multi_subgraphs.md b/docs/serving/docs/source_en/serving_multi_subgraphs.md index 56e7a42bd27550fe3d9adcf01c26c5445300fe25..026c9c55a5d1262d8ebbc9dbc700f16928794034 100644 --- a/docs/serving/docs/source_en/serving_multi_subgraphs.md +++ b/docs/serving/docs/source_en/serving_multi_subgraphs.md @@ -8,7 +8,7 @@ MindSpore allows a model to generate multiple subgraphs. Such a model is general MindSpore Serving supports scheduling multi-subgraph model, improving the inference service performance in specific scenarios. -The following uses a simple single-chip model scenario as an example to describe the multi-subgraph model deployment process. For the detail about the distributed scenario, see [Pengcheng·Pangu Model model Serving deployment](https://gitee.com/mindspore/models/tree/master/official/nlp/pangu_alpha#serving) +The following uses a simple single-chip model scenario as an example to describe the multi-subgraph model deployment process. For the detail about the distributed scenario, see [Pengcheng·Pangu Model model Serving deployment](https://gitee.com/mindspore/models/tree/r1.7/official/nlp/pangu_alpha#serving) ### Environment Preparation diff --git a/docs/serving/docs/source_zh_cn/serving_distributed_example.md b/docs/serving/docs/source_zh_cn/serving_distributed_example.md index 9c70c8247789be8ed6cdfe67a853c51e21d8cb31..fa0f3133bf101ee11e6ae4baeec1914ec7203ced 100644 --- a/docs/serving/docs/source_zh_cn/serving_distributed_example.md +++ b/docs/serving/docs/source_zh_cn/serving_distributed_example.md @@ -42,7 +42,7 @@ export_model - `net.py`为MatMul网络定义。 - `distributed_inference.py`配置分布式相关的参数。 - `export_model.sh`在当前机器上创建`device`目录并且导出每个`device`对应的模型文件。 -- `rank_table_8pcs.json`为配置当前多卡环境的组网信息的json文件，可以参考[rank_table](https://gitee.com/mindspore/models/tree/master/utils/hccl_tools)。 +- `rank_table_8pcs.json`为配置当前多卡环境的组网信息的json文件，可以参考[rank_table](https://gitee.com/mindspore/models/tree/r1.7/utils/hccl_tools)。 使用[net.py](https://gitee.com/mindspore/serving/blob/r1.7/example/matmul_distributed/export_model/net.py)，构造一个包含MatMul、Neg算子的网络。 diff --git a/docs/serving/docs/source_zh_cn/serving_multi_subgraphs.md b/docs/serving/docs/source_zh_cn/serving_multi_subgraphs.md index ab0b5547344ac1f8a02af889af1403577bf2b8cf..18b978eeb20dc394b46d30c695e1454687c98780 100644 --- a/docs/serving/docs/source_zh_cn/serving_multi_subgraphs.md +++ b/docs/serving/docs/source_zh_cn/serving_multi_subgraphs.md @@ -6,7 +6,7 @@ MindSpore支持一个模型导出生成多张子图，拥有多个子图的模型一般也是有状态的模型，多个子图之间共享权重，通过多个子图配合实现性能优化等目标。例如，在鹏程·盘古模型网络场景，基于一段语句，经过多次推理产生一段语句，其中每次推理产生一个词。不同输入长度将会产生两个图，第一为输入长度为1024的全量输入图，处理首次长度不定文本，只需执行一次，第二图为输入长度为1的增量输入图，处理上一次新增的字，第二个图将执行多次。相对于优化之前仅有全量图执行多次，可实现推理服务性能的5-6倍提升。为此，MindSpore Serving提供了多子图功能，实现多张图之间的调度。 -下面以一个简单的单卡模型场景为例，演示多子图模型部署流程，分布式场景可以参考[鹏程·盘古模型模型Serving部署](https://gitee.com/mindspore/models/tree/master/official/nlp/pangu_alpha#serving)。 +下面以一个简单的单卡模型场景为例，演示多子图模型部署流程，分布式场景可以参考[鹏程·盘古模型模型Serving部署](https://gitee.com/mindspore/models/tree/r1.7/official/nlp/pangu_alpha#serving)。 ### 环境准备 diff --git a/install/mindspore_ascend310_install_conda.md b/install/mindspore_ascend310_install_conda.md index 7ea503ce40b23ccfb9ab48af4a58e394c2eb01cd..fa913122e445477ce0dcbe8092d4474365c6cdc7 100644 --- a/install/mindspore_ascend310_install_conda.md +++ b/install/mindspore_ascend310_install_conda.md @@ -177,7 +177,7 @@ pip uninstall te topi hccl -y conda install mindspore-ascend -c mindspore -c conda-forge ``` -在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 配置环境变量 diff --git a/install/mindspore_ascend310_install_conda_en.md b/install/mindspore_ascend310_install_conda_en.md index 43d4ce45e3ba0e501dd52b30b2b03858c50018a3..b0ae90b002595e4b2b2386b052c08ce5441975e0 100644 --- a/install/mindspore_ascend310_install_conda_en.md +++ b/install/mindspore_ascend310_install_conda_en.md @@ -177,7 +177,7 @@ Ensure that you are in the Conda virtual environment and run the following comma conda install mindspore-ascend -c mindspore -c conda-forge ``` -When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Configuring Environment Variables diff --git a/install/mindspore_ascend310_install_pip.md b/install/mindspore_ascend310_install_pip.md index 79a09f7f10ec4689fdc23c542411936e2b56146d..0cd572e9fe59f53b588f65c5c5dce4741969bda8 100644 --- a/install/mindspore_ascend310_install_pip.md +++ b/install/mindspore_ascend310_install_pip.md @@ -236,7 +236,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/aarch64/mindspore_ascend-${MS_VERSION/-/}-cp39-cp39-linux_aarch64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装whl包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装whl包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 配置环境变量 diff --git a/install/mindspore_ascend310_install_pip_en.md b/install/mindspore_ascend310_install_pip_en.md index 9862b12651867a78c6549e8256b79440ebf5b0f4..69636c03d7cf4695be284efdf6148253885c2650 100644 --- a/install/mindspore_ascend310_install_pip_en.md +++ b/install/mindspore_ascend310_install_pip_en.md @@ -236,7 +236,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/aarch64/mindspore_ascend-${MS_VERSION/-/}-cp39-cp39-linux_aarch64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. For details about dependencies, see required_package in the [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py). In other cases, install the dependencies by yourself. When running a model, you need to install additional dependencies based on the requirements.txt file specified by different models in the [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. For details about dependencies, see required_package in the [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py). In other cases, install the dependencies by yourself. When running a model, you need to install additional dependencies based on the requirements.txt file specified by different models in the [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Configuring Environment Variables diff --git a/install/mindspore_ascend_install_conda.md b/install/mindspore_ascend_install_conda.md index 33c38cd73fe37330b562a99cf207525b467f2b5f..0211d500c54fddbc715868b3b0fc24b885531d85 100644 --- a/install/mindspore_ascend_install_conda.md +++ b/install/mindspore_ascend_install_conda.md @@ -198,7 +198,7 @@ pip uninstall te topi hccl -y conda install mindspore-ascend -c mindspore -c conda-forge ``` -在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 配置环境变量 diff --git a/install/mindspore_ascend_install_conda_en.md b/install/mindspore_ascend_install_conda_en.md index 433f4f316332caac9dbe9d02f4089120efefce9a..0f9e302c4d80bf2b0a4d1cce195c83e1a186286e 100644 --- a/install/mindspore_ascend_install_conda_en.md +++ b/install/mindspore_ascend_install_conda_en.md @@ -198,7 +198,7 @@ Ensure that you are in the Conda virtual environment and run the following comma conda install mindspore-ascend -c mindspore -c conda-forge ``` -When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Configuring Environment Variables diff --git a/install/mindspore_ascend_install_pip.md b/install/mindspore_ascend_install_pip.md index 01bc8dd852ecb37e75d1b527120545fe50125ee5..0fc23597c186652cbe2e877375a66c68d6f1f0a5 100644 --- a/install/mindspore_ascend_install_pip.md +++ b/install/mindspore_ascend_install_pip.md @@ -220,7 +220,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/aarch64/mindspore_ascend-${MS_VERSION/-/}-cp39-cp39-linux_aarch64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装whl包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装whl包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 配置环境变量 diff --git a/install/mindspore_ascend_install_pip_en.md b/install/mindspore_ascend_install_pip_en.md index 72e1aa31bcb50f43b8ec21ffce667f46bb7454a5..59c2b6b5b8b32b23d095fc1bf6ddde69ff9d8e0b 100644 --- a/install/mindspore_ascend_install_pip_en.md +++ b/install/mindspore_ascend_install_pip_en.md @@ -220,7 +220,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/aarch64/mindspore_ascend-${MS_VERSION/-/}-cp39-cp39-linux_aarch64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Configuring Environment Variables diff --git a/install/mindspore_ascend_install_source.md b/install/mindspore_ascend_install_source.md index af3986ca22926f8b1a8a9e676563dd25fc860a33..581f0abe2d11b74f2fd4dcc463f32524d95b80cb 100644 --- a/install/mindspore_ascend_install_source.md +++ b/install/mindspore_ascend_install_source.md @@ -311,7 +311,7 @@ bash build.sh -e ascend -S on pip install output/mindspore_ascend-*.whl -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装whl包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装whl包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 配置环境变量 diff --git a/install/mindspore_ascend_install_source_en.md b/install/mindspore_ascend_install_source_en.md index 82fdded4a5bb5f930f75a1da6fe20cb0a98c013a..2298e8a5039346953715401aaaf3ae1c4a398732 100644 --- a/install/mindspore_ascend_install_source_en.md +++ b/install/mindspore_ascend_install_source_en.md @@ -311,7 +311,7 @@ Where: pip install output/mindspore_ascend-*.whl -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Configuring Environment Variables diff --git a/install/mindspore_cpu_install_conda.md b/install/mindspore_cpu_install_conda.md index 7fb28facdf4109c7071d059e9053372d68f1119c..12d54a0c42823d7fee57aea0589e114adb63fcf3 100644 --- a/install/mindspore_cpu_install_conda.md +++ b/install/mindspore_cpu_install_conda.md @@ -129,7 +129,7 @@ conda activate mindspore_py37 conda install mindspore-cpu -c mindspore -c conda-forge -y ``` -在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_install_conda_en.md b/install/mindspore_cpu_install_conda_en.md index 93819127049765e27dafc852067a9ae240a5864b..a320f7180777cb97a94f87c8ea516f9e4a822f8f 100644 --- a/install/mindspore_cpu_install_conda_en.md +++ b/install/mindspore_cpu_install_conda_en.md @@ -129,7 +129,7 @@ Ensure that you are in the Conda virtual environment and run the following comma conda install mindspore-cpu -c mindspore -c conda-forge ``` -When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_cpu_install_nightly.md b/install/mindspore_cpu_install_nightly.md index 087876b56c8f2e8f118757b3c4286d074aa6221f..a9b459d32dc7c3627b958b7fbf6e96797ac9c91b 100644 --- a/install/mindspore_cpu_install_nightly.md +++ b/install/mindspore_cpu_install_nightly.md @@ -115,7 +115,7 @@ pip install mindspore-dev -i https://pypi.tuna.tsinghua.edu.cn/simple 其中： -- 在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +- 在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 - pip会自动安装当前最新版本的Nightly版本MindSpore，如果需要安装指定版本，请参照下方升级MindSpore版本相关指导，在下载时手动指定版本。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_install_nightly_en.md b/install/mindspore_cpu_install_nightly_en.md index 5c28c04471d08208d1c2f97c555663182823afa9..3e81db8720f6f04a92278a8ed6dff908a1e2350c 100644 --- a/install/mindspore_cpu_install_nightly_en.md +++ b/install/mindspore_cpu_install_nightly_en.md @@ -115,7 +115,7 @@ pip install mindspore-dev -i https://pypi.tuna.tsinghua.edu.cn/simple Of which, -- When the network is connected, dependencies are automatically downloaded during .whl package installation. (For details about the dependencies, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +- When the network is connected, dependencies are automatically downloaded during .whl package installation. (For details about the dependencies, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). - pip will be installing the latest version of MindSpore Nightly automatically. If you wish to specify the version to be installed, please refer to the instruction below regarding to version update, and specify version manually. ## Installation Verification diff --git a/install/mindspore_cpu_install_pip.md b/install/mindspore_cpu_install_pip.md index 5fb7a95fc7eac476e481e0323f112b5132130d88..7aee25620e7ce7bf1dbfc76dcb17094546d7ecfd 100644 --- a/install/mindspore_cpu_install_pip.md +++ b/install/mindspore_cpu_install_pip.md @@ -154,7 +154,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/cpu/aarch64/mindspore-${MS_VERSION/-/}-cp39-cp39-linux_aarch64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_install_pip_en.md b/install/mindspore_cpu_install_pip_en.md index 2a82c23fcf02032a62b28c448ab12805d8b9de13..645a8a45ee813d5138aac212e56c5370757522da 100644 --- a/install/mindspore_cpu_install_pip_en.md +++ b/install/mindspore_cpu_install_pip_en.md @@ -154,7 +154,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/cpu/aarch64/mindspore-${MS_VERSION/-/}-cp39-cp39-linux_aarch64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_cpu_install_source.md b/install/mindspore_cpu_install_source.md index 8cd3791e35938f104111858fa563f99ae295d641..43cdf04550dfa06d255cb148e78d494611b44a8b 100644 --- a/install/mindspore_cpu_install_source.md +++ b/install/mindspore_cpu_install_source.md @@ -198,7 +198,7 @@ bash build.sh -e cpu -j4 -S on pip install output/mindspore-*.whl -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证安装是否成功 diff --git a/install/mindspore_cpu_install_source_en.md b/install/mindspore_cpu_install_source_en.md index fcf044aa09648c3edf6ed36185e5d007033e1b9a..cd3c3b6bb9e3403ac8de30e3a16e71e83a703ba2 100644 --- a/install/mindspore_cpu_install_source_en.md +++ b/install/mindspore_cpu_install_source_en.md @@ -196,7 +196,7 @@ Where: pip install output/mindspore-*.whl -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. For details about dependencies, see required_package in the [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py). In other cases, install the dependencies by yourself. When running a model, you need to install additional dependencies based on the requirements.txt file specified by different models in the [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. For details about dependencies, see required_package in the [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py). In other cases, install the dependencies by yourself. When running a model, you need to install additional dependencies based on the requirements.txt file specified by different models in the [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_cpu_mac_install_conda.md b/install/mindspore_cpu_mac_install_conda.md index eae7ca4b11b9f54d46ec803891b3433270a23fe1..85016b953118c3639ad9f4da7c2658ae028250d4 100644 --- a/install/mindspore_cpu_mac_install_conda.md +++ b/install/mindspore_cpu_mac_install_conda.md @@ -64,7 +64,7 @@ conda install mindspore-cpu -c mindspore -c conda-forge ``` -在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_mac_install_conda_en.md b/install/mindspore_cpu_mac_install_conda_en.md index 25012afe59230d16c0e7c11b1f276f75a6e77fe2..338e725c04e970dd32d2977e245451f51fc244f1 100644 --- a/install/mindspore_cpu_mac_install_conda_en.md +++ b/install/mindspore_cpu_mac_install_conda_en.md @@ -64,7 +64,7 @@ Ensure that you are in the Conda virtual environment and run the following comma conda install mindspore-cpu -c mindspore -c conda-forge ``` -When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_cpu_mac_install_nightly.md b/install/mindspore_cpu_mac_install_nightly.md index 63ebd4c79f39b9849b49c28ae9430d7e21daee36..662e97463a68ec53a8a4e3f3939e3e6eeb1716fd 100644 --- a/install/mindspore_cpu_mac_install_nightly.md +++ b/install/mindspore_cpu_mac_install_nightly.md @@ -41,7 +41,7 @@ pip install mindspore-dev -i https://pypi.tuna.tsinghua.edu.cn/simple 其中： -- 在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +- 在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 - pip会自动安装当前最新版本的Nightly版本MindSpore，如果需要安装指定版本，请参照下方升级MindSpore版本相关指导，在下载时手动指定版本。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_mac_install_nightly_en.md b/install/mindspore_cpu_mac_install_nightly_en.md index ceb6e4cc96a7fb3d60c0d9a85f5e16a1783d25dc..a4dc05baae4b38de142e659da3abf56009bde3cb 100644 --- a/install/mindspore_cpu_mac_install_nightly_en.md +++ b/install/mindspore_cpu_mac_install_nightly_en.md @@ -41,7 +41,7 @@ pip install mindspore-dev -i https://pypi.tuna.tsinghua.edu.cn/simple Of which, -- When the network is connected, dependencies are automatically downloaded during .whl package installation. (For details about the dependencies, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +- When the network is connected, dependencies are automatically downloaded during .whl package installation. (For details about the dependencies, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). - pip will be installing the latest version of MindSpore Nightly automatically. If you wish to specify the version to be installed, please refer to the instruction below regarding to version update, and specify version manually. ## Installation Verification diff --git a/install/mindspore_cpu_mac_install_pip.md b/install/mindspore_cpu_mac_install_pip.md index b95045e9ff407ee0ab8131500385c0d1c60434be..15bbec69c43202913956074b81edfa4e8f151611 100644 --- a/install/mindspore_cpu_mac_install_pip.md +++ b/install/mindspore_cpu_mac_install_pip.md @@ -52,7 +52,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/cpu/aarch64/mindspore-${MS_VERSION/-/}-cp39-cp39-macosx_11_0_arm64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_mac_install_pip_en.md b/install/mindspore_cpu_mac_install_pip_en.md index f9ca8848f7e1b54953183049a831b6905d872847..caab6b44e601ce487385219eebc88529b49db739 100644 --- a/install/mindspore_cpu_mac_install_pip_en.md +++ b/install/mindspore_cpu_mac_install_pip_en.md @@ -52,7 +52,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/cpu/aarch64/mindspore-${MS_VERSION/-/}-cp39-cp39-macosx_11_0_arm64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. For details about dependencies, see required_package in the [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py). In other cases, install the dependencies by yourself. When running a model, you need to install additional dependencies based on the requirements.txt file specified by different models in the [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. For details about dependencies, see required_package in the [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py). In other cases, install the dependencies by yourself. When running a model, you need to install additional dependencies based on the requirements.txt file specified by different models in the [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_cpu_win_install_conda.md b/install/mindspore_cpu_win_install_conda.md index e071fb65e56b556a315ee2d0b2b84d509f026327..1bc4f9a2560ca6e8f5a4f8889a52eadf9cede9ce 100644 --- a/install/mindspore_cpu_win_install_conda.md +++ b/install/mindspore_cpu_win_install_conda.md @@ -59,7 +59,7 @@ activate mindspore_py39 conda install mindspore-cpu -c mindspore -c conda-forge ``` -在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装Conda安装包时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_win_install_conda_en.md b/install/mindspore_cpu_win_install_conda_en.md index 64ff47acde3131967b2e266fe1c4cb38c194adb2..7311d01af68923bbd592026818f4c2cdcbf46fee 100644 --- a/install/mindspore_cpu_win_install_conda_en.md +++ b/install/mindspore_cpu_win_install_conda_en.md @@ -59,7 +59,7 @@ Ensure that you are in the Conda virtual environment and run the following comma conda install mindspore-cpu -c mindspore -c conda-forge ``` -When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_cpu_win_install_nightly.md b/install/mindspore_cpu_win_install_nightly.md index d208b016bbbb328777a77c463665ddc6ae8cf43d..d2601e7e2aa4824cf42db0e438cde784133fa972 100644 --- a/install/mindspore_cpu_win_install_nightly.md +++ b/install/mindspore_cpu_win_install_nightly.md @@ -39,7 +39,7 @@ pip install mindspore-dev -i https://pypi.tuna.tsinghua.edu.cn/simple 其中： -- 在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +- 在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 - pip会自动安装当前最新版本的Nightly版本MindSpore，如果需要安装指定版本，请参照下方升级MindSpore版本相关指导，在下载时手动指定版本。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_win_install_nightly_en.md b/install/mindspore_cpu_win_install_nightly_en.md index bc99f71258a9294e7db9a5e97d272236af1a1aab..ba13b09f2d0aed82a6a2e816c4b1f6194bdba502 100644 --- a/install/mindspore_cpu_win_install_nightly_en.md +++ b/install/mindspore_cpu_win_install_nightly_en.md @@ -39,7 +39,7 @@ pip install mindspore-dev -i https://pypi.tuna.tsinghua.edu.cn/simple Of which, -- When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +- When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). - pip will be installing the latest version of MindSpore Nightly automatically. If you wish to specify the version to be installed, please refer to the instruction below regarding to version update, and specify version manually. ## Installation Verification diff --git a/install/mindspore_cpu_win_install_pip.md b/install/mindspore_cpu_win_install_pip.md index 8005507387b119190c33c79468db318171f8478f..eef72890e07344e2bc45c34a740934dac1532b6d 100644 --- a/install/mindspore_cpu_win_install_pip.md +++ b/install/mindspore_cpu_win_install_pip.md @@ -46,7 +46,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/%MS_VERSION%/Min pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/%MS_VERSION%/MindSpore/cpu/x86_64/mindspore-%MS_VERSION:-=%-cp39-cp39-win_amd64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_win_install_pip_en.md b/install/mindspore_cpu_win_install_pip_en.md index 6d76db3af8e744bf274e79bd72ed97a92c0d2fde..5a3bf17c0aec9bd527a027686cffba415af35b15 100644 --- a/install/mindspore_cpu_win_install_pip_en.md +++ b/install/mindspore_cpu_win_install_pip_en.md @@ -46,7 +46,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/%MS_VERSION%/Min pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/%MS_VERSION%/MindSpore/cpu/x86_64/mindspore-%MS_VERSION:-=%-cp39-cp39-win_amd64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_cpu_win_install_source.md b/install/mindspore_cpu_win_install_source.md index f3524f308c35e01a13942d5bbbdd120ac6400683..1e37c8d2977530ef9dcf4c218ac3512c242bcbdc 100644 --- a/install/mindspore_cpu_win_install_source.md +++ b/install/mindspore_cpu_win_install_source.md @@ -56,7 +56,7 @@ call build.bat for %x in (output\mindspore*.whl) do pip install %x -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_cpu_win_install_source_en.md b/install/mindspore_cpu_win_install_source_en.md index 1c7e4375a380b51eefe93f16d0e1086fb2f2fe2e..ea3db02aa8517a5cd07ba61ea6650276e9546c39 100644 --- a/install/mindspore_cpu_win_install_source_en.md +++ b/install/mindspore_cpu_win_install_source_en.md @@ -56,7 +56,7 @@ call build.bat for %x in (output\mindspore*.whl) do pip install %x -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_gpu_install_conda.md b/install/mindspore_gpu_install_conda.md index c969e18ebc11e0f8c793d9f7890c3fbb551a59b4..9bfa550e2cbab1407b25c23e1f8a716aa1bd2a70 100644 --- a/install/mindspore_gpu_install_conda.md +++ b/install/mindspore_gpu_install_conda.md @@ -238,7 +238,7 @@ CUDA 11.1版本： conda install mindspore-gpu cudatoolkit=11.1 -c mindspore -c conda-forge ``` -在联网状态下，安装MindSpore时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装MindSpore时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_gpu_install_conda_en.md b/install/mindspore_gpu_install_conda_en.md index aad9b7bafc70bb45bb1882be61c6897f913f2722..ad3ff2fc45ce3f69d21fbc384da154bb4d397b8d 100644 --- a/install/mindspore_gpu_install_conda_en.md +++ b/install/mindspore_gpu_install_conda_en.md @@ -238,7 +238,7 @@ For CUDA 11.1: conda install mindspore-gpu cudatoolkit=11.1 -c mindspore -c conda-forge ``` -When the network is connected, dependency items are automatically downloaded during MindSpore installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependency items are automatically downloaded during MindSpore installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_gpu_install_nightly.md b/install/mindspore_gpu_install_nightly.md index be90c30d08f70692d2c7e3a7860197182e99814e..f622565130a64bda17a520a324f67da6880c571f 100644 --- a/install/mindspore_gpu_install_nightly.md +++ b/install/mindspore_gpu_install_nightly.md @@ -200,7 +200,7 @@ pip install mindspore-cuda11-dev -i https://pypi.tuna.tsinghua.edu.cn/simple 其中： - 当前MindSpore GPU Nightly仅提供CUDA11版本。 -- 在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +- 在联网状态下，安装whl包时会自动下载mindspore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 - pip会自动安装当前最新版本的Nightly版本MindSpore，如果需要安装指定版本，请参照下方升级MindSpore版本相关指导，在下载时手动指定版本。 ## 验证是否成功安装 diff --git a/install/mindspore_gpu_install_nightly_en.md b/install/mindspore_gpu_install_nightly_en.md index 98292f2829cae6bf790e95976b8d3649043081b5..00b6a02bd5c3922280a75465914dc8b4537a39c7 100644 --- a/install/mindspore_gpu_install_nightly_en.md +++ b/install/mindspore_gpu_install_nightly_en.md @@ -200,7 +200,7 @@ pip install mindspore-cuda11-dev -i https://pypi.tuna.tsinghua.edu.cn/simple Of which, - Currently, MindSpore GPU Nightly only provides CUDA11 version. -- When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +- When the network is connected, dependency items are automatically downloaded during .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). - pip will be installing the latest version of MindSpore GPU Nightly automatically. If you wish to specify the version to be installed, please refer to the instruction below regarding to version update, and specify version manually. ## Installation Verification diff --git a/install/mindspore_gpu_install_pip.md b/install/mindspore_gpu_install_pip.md index c0ea103caca428ea863d607b19ab389d74c3b849..8bc926b5cc261813038cbf0e9a7e453a0c1e83b2 100644 --- a/install/mindspore_gpu_install_pip.md +++ b/install/mindspore_gpu_install_pip.md @@ -256,7 +256,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/gpu/x86_64/cuda-11.1/mindspore_gpu-${MS_VERSION/-/}-cp39-cp39-linux_x86_64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装MindSpore时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装MindSpore时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_gpu_install_pip_en.md b/install/mindspore_gpu_install_pip_en.md index fac4ba62949b49dbc14817a5480eab469e3b6ee4..742c646b2e7b2705d89d53160af1c3375d060efb 100644 --- a/install/mindspore_gpu_install_pip_en.md +++ b/install/mindspore_gpu_install_pip_en.md @@ -256,7 +256,7 @@ pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/Mi pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/gpu/x86_64/cuda-11.1/mindspore_gpu-${MS_VERSION/-/}-cp39-cp39-linux_x86_64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependency items are automatically downloaded during MindSpore installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependency items are automatically downloaded during MindSpore installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/install/mindspore_gpu_install_source.md b/install/mindspore_gpu_install_source.md index 87b0286083b72a264ef5ae50cfa8e8ecaf901bfc..ddb49fb9ee47da3bb0f79316af5ac7a86ff953fd 100644 --- a/install/mindspore_gpu_install_source.md +++ b/install/mindspore_gpu_install_source.md @@ -314,7 +314,7 @@ bash build.sh -e gpu -S on pip install output/mindspore_gpu-*.whl -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -在联网状态下，安装MindSpore时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/master/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 +在联网状态下，安装MindSpore时会自动下载MindSpore安装包的依赖项（依赖项详情参见[setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py)中的required_package），其余情况需自行安装。运行模型时，需要根据[ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/)中不同模型指定的requirements.txt安装额外依赖，常见依赖可以参考[requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt)。 ## 验证是否成功安装 diff --git a/install/mindspore_gpu_install_source_en.md b/install/mindspore_gpu_install_source_en.md index 60ede147b485d66551ce41e0c01603106c87c524..2c1b746ec696d52fd09a3deec297ea0e26764d10 100644 --- a/install/mindspore_gpu_install_source_en.md +++ b/install/mindspore_gpu_install_source_en.md @@ -314,7 +314,7 @@ Where: pip install output/mindspore_gpu-*.whl -i https://pypi.tuna.tsinghua.edu.cn/simple ``` -When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/master/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). +When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in [setup.py](https://gitee.com/mindspore/mindspore/blob/r1.7/setup.py) .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in [ModelZoo](https://gitee.com/mindspore/models/tree/r1.7/). For details about common dependencies, see [requirements.txt](https://gitee.com/mindspore/mindspore/blob/r1.7/requirements.txt). ## Installation Verification diff --git a/tutorials/experts/source_en/debug/dataset_autotune.md b/tutorials/experts/source_en/debug/dataset_autotune.md index 6657ee9c729a1ed31746f06bf9a2d561924a5d83..a754a6d6c24576c932131874b7740d73956452ce 100644 --- a/tutorials/experts/source_en/debug/dataset_autotune.md +++ b/tutorials/experts/source_en/debug/dataset_autotune.md @@ -116,7 +116,7 @@ def create_dataset(...) ### Start Training -Start the training process as described in [resnet/README.md](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/README.md#). Dataset AutoTune will display its analysis result through LOG messages. +Start the training process as described in [resnet/README.md](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/README.md#). Dataset AutoTune will display its analysis result through LOG messages. ```text [INFO] [auto_tune.cc:73 LaunchThread] Launching Dataset AutoTune thread diff --git a/tutorials/experts/source_en/debug/dump.md b/tutorials/experts/source_en/debug/dump.md index fdd89c869e9dff8c81f522317812abd5a72b4f08..d4d5af4c5aca900695fd208f208c785418556db9 100644 --- a/tutorials/experts/source_en/debug/dump.md +++ b/tutorials/experts/source_en/debug/dump.md @@ -235,7 +235,7 @@ Since sub-graphs share the same graph execution history with root graph, only ro For the Ascend scene, after the graph corresponding to the script is saved to the disk through the Dump function, the final execution graph file `ms_output_trace_code_graph_{graph_id}.ir` will be generated. This file saves the stack information of each operator in the corresponding graph, and records the generation script corresponding to the operator. -Take [AlexNet script](https://gitee.com/mindspore/models/blob/master/official/cv/alexnet/src/alexnet.py) as an example: +Take [AlexNet script](https://gitee.com/mindspore/models/blob/r1.7/official/cv/alexnet/src/alexnet.py) as an example: ```python import mindspore.nn as nn diff --git a/tutorials/experts/source_en/infer/cpu_gpu_mindir.md b/tutorials/experts/source_en/infer/cpu_gpu_mindir.md index 26381e1400695d6525dc2f50d0a94568e840b02f..13fc91ca9a201647cb171fd6436fb3e8a558d69a 100644 --- a/tutorials/experts/source_en/infer/cpu_gpu_mindir.md +++ b/tutorials/experts/source_en/infer/cpu_gpu_mindir.md @@ -150,7 +150,7 @@ infer finished. ### Notices -- During the training process, some networks set operator precision to FP16 artificially. For example, the [Bert mode](https://gitee.com/mindspore/models/blob/master/official/nlp/bert/src/bert_model.py) set the `Dense` and `LayerNorm` to FP16: +- During the training process, some networks set operator precision to FP16 artificially. For example, the [Bert mode](https://gitee.com/mindspore/models/blob/r1.7/official/nlp/bert/src/bert_model.py) set the `Dense` and `LayerNorm` to FP16: ```python class BertOutput(nn.Cell): diff --git a/tutorials/experts/source_en/others/gradient_accumulation.md b/tutorials/experts/source_en/others/gradient_accumulation.md index ac401c05abbdce061eb146bbae2e65b38f2e4104..8242e8620f40054aa200ecef3300a3d5aadbaf81 100644 --- a/tutorials/experts/source_en/others/gradient_accumulation.md +++ b/tutorials/experts/source_en/others/gradient_accumulation.md @@ -53,11 +53,11 @@ from models.official.cv.lenet.src.lenet import LeNet5 ### Loading the Dataset -Use the `MnistDataset` API provided by `dataset` of MindSpore to load the MNIST dataset. The code is imported from [dataset.py](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/src/dataset.py) in the `lenet` directory of `models`. +Use the `MnistDataset` API provided by `dataset` of MindSpore to load the MNIST dataset. The code is imported from [dataset.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/src/dataset.py) in the `lenet` directory of `models`. ### Defining the Network -LeNet is used as an example network. You can also use other networks, such as ResNet-50 and BERT. The code is imported from [lenet.py](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/src/lenet.py) in the `lenet` directory of `models`. +LeNet is used as an example network. You can also use other networks, such as ResNet-50 and BERT. The code is imported from [lenet.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/src/lenet.py) in the `lenet` directory of `models`. ### Defining the Training Process @@ -255,7 +255,7 @@ After 10 epochs, the accuracy on the test set is about 96.31%. **Validate the model.** -Use the saved checkpoint file to load the validation dataset through [eval.py]() in the lenet directory of models. +Use the saved checkpoint file to load the validation dataset through [eval.py]() in the lenet directory of models. ```bash python eval.py --data_path=./MNIST_Data --ckpt_path=./gradient_accumulation.ckpt --device_target=GPU diff --git a/tutorials/experts/source_zh_cn/dataset/cache.ipynb b/tutorials/experts/source_zh_cn/dataset/cache.ipynb index 4691b773d35fdc1eac42fe8191f054cfc9d55128..29523d0de66ff204ae8bbf9bee6ffda1b0259e07 100644 --- a/tutorials/experts/source_zh_cn/dataset/cache.ipynb +++ b/tutorials/experts/source_zh_cn/dataset/cache.ipynb @@ -783,7 +783,7 @@ "\n", "因此我们可以选择直接在数据集读取算子之后添加缓存。本节将采用这种方法，以MobileNetV2网络为样本，进行示例。\n", "\n", - "完整示例代码请参考ModelZoo的[MobileNetV2](https://gitee.com/mindspore/models/tree/master/official/cv/mobilenetv2)。\n", + "完整示例代码请参考ModelZoo的[MobileNetV2](https://gitee.com/mindspore/models/tree/r1.7/official/cv/mobilenetv2)。\n", "\n", "1. 创建管理缓存的Shell脚本`cache_util.sh`：\n", "\n", diff --git a/tutorials/experts/source_zh_cn/debug/dataset_autotune.md b/tutorials/experts/source_zh_cn/debug/dataset_autotune.md index 8051e89f681f0bc389a3fc26f556906e33296116..8f2ea9cc56f83d44cf5eb927c17451fa71e6a1d9 100644 --- a/tutorials/experts/source_zh_cn/debug/dataset_autotune.md +++ b/tutorials/experts/source_zh_cn/debug/dataset_autotune.md @@ -115,7 +115,7 @@ def create_dataset(...) ### 开始训练 -根据[resnet/README.md](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/README_CN.md#)所描述的步骤 +根据[resnet/README.md](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/README_CN.md#)所描述的步骤 启动CIFAR10数据集的训练，随后自动数据加速模块会通过LOG的形式展示其对于性能瓶颈的分析情况： ```text diff --git a/tutorials/experts/source_zh_cn/debug/dump.md b/tutorials/experts/source_zh_cn/debug/dump.md index 8741111fd6b980e0d86c64d6edcdbc54e563d099..09da41f8dc64e87c4b7b7d89a0e2e8780e4e7592 100644 --- a/tutorials/experts/source_zh_cn/debug/dump.md +++ b/tutorials/experts/source_zh_cn/debug/dump.md @@ -235,7 +235,7 @@ ms_global_execution_order_graph_{graph_id}.csv 对于Ascend场景，在通过Dump功能将脚本对应的图保存到磁盘上后，会产生最终执行图文件`ms_output_trace_code_graph_{graph_id}.ir`。该文件中保存了对应的图中每个算子的堆栈信息，记录了算子对应的生成脚本。 -以[AlexNet脚本](https://gitee.com/mindspore/models/blob/master/official/cv/alexnet/src/alexnet.py)为例 ： +以[AlexNet脚本](https://gitee.com/mindspore/models/blob/r1.7/official/cv/alexnet/src/alexnet.py)为例 ： ```python import mindspore.nn as nn diff --git a/tutorials/experts/source_zh_cn/infer/cpu_gpu_mindir.md b/tutorials/experts/source_zh_cn/infer/cpu_gpu_mindir.md index 324b49fd7fac8b31fe7e3c8d7a70ad8bc52b106d..bd354059b0f0f7003938a8ee82fa42ab0db391c9 100644 --- a/tutorials/experts/source_zh_cn/infer/cpu_gpu_mindir.md +++ b/tutorials/experts/source_zh_cn/infer/cpu_gpu_mindir.md @@ -165,7 +165,7 @@ infer finished. ### 备注 -- 一些网络在训练过程时，人为将部分算子精度设置为FP16。例如ModelZoo中的[Bert网络](https://gitee.com/mindspore/models/blob/master/official/nlp/bert/src/bert_model.py)，将Dense和LayerNorm设置为FP16进行训练。 +- 一些网络在训练过程时，人为将部分算子精度设置为FP16。例如ModelZoo中的[Bert网络](https://gitee.com/mindspore/models/blob/r1.7/official/nlp/bert/src/bert_model.py)，将Dense和LayerNorm设置为FP16进行训练。 ```python class BertOutput(nn.Cell): diff --git a/tutorials/experts/source_zh_cn/infer/inference.md b/tutorials/experts/source_zh_cn/infer/inference.md index bdf747fd0b40e0eea299bb11f011f2737a684712..10ebaa5ff0de7b4bddcec7824f3540a08931e8ec 100644 --- a/tutorials/experts/source_zh_cn/infer/inference.md +++ b/tutorials/experts/source_zh_cn/infer/inference.md @@ -85,7 +85,7 @@ print("============== {} ==============".format(acc)) 其中， `model.eval`为模型验证接口，对应接口说明：。 -> 推理样例代码：。 +> 推理样例代码：。 ### 使用MindSpore Hub从华为云加载模型 diff --git a/tutorials/experts/source_zh_cn/others/adaptive_summation.md b/tutorials/experts/source_zh_cn/others/adaptive_summation.md index d8cdc1edd4e0ad47603984f37c71fec3198de6c3..85025ee01f012443744852e9eb9d8bb9f63d988c 100644 --- a/tutorials/experts/source_zh_cn/others/adaptive_summation.md +++ b/tutorials/experts/source_zh_cn/others/adaptive_summation.md @@ -95,7 +95,7 @@ $$ } ``` -rank_table可以使用models下面的[hccl_tools.py](https://gitee.com/mindspore/models/blob/master/utils/hccl_tools/hccl_tools.py)生成，[merge_hccl.py](https://gitee.com/mindspore/models/blob/master/utils/hccl_tools/merge_hccl.py)可将多个rank_table文件进行拼接。脚本使用方法可见[README.md](https://gitee.com/mindspore/models/blob/master/utils/hccl_tools/README.md#)。 +rank_table可以使用models下面的[hccl_tools.py](https://gitee.com/mindspore/models/blob/r1.7/utils/hccl_tools/hccl_tools.py)生成，[merge_hccl.py](https://gitee.com/mindspore/models/blob/r1.7/utils/hccl_tools/merge_hccl.py)可将多个rank_table文件进行拼接。脚本使用方法可见[README.md](https://gitee.com/mindspore/models/blob/r1.7/utils/hccl_tools/README.md#)。 ### 数据集准备 @@ -128,7 +128,7 @@ init() ## 数据并行模式加载数据集 -分布式训练时，数据以数据并行的方式导入。利用MindSpore提供图片加载接口ImageFolderDataset加载ImageNet 2012数据集，同时通过MindSpore提供的数据增强接口对数据集进行处理，此部分代码由models中`resnet`目录下的[dataset.py](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/src/dataset.py)导入。 +分布式训练时，数据以数据并行的方式导入。利用MindSpore提供图片加载接口ImageFolderDataset加载ImageNet 2012数据集，同时通过MindSpore提供的数据增强接口对数据集进行处理，此部分代码由models中`resnet`目录下的[dataset.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/src/dataset.py)导入。 ```python # define train dataset diff --git a/tutorials/experts/source_zh_cn/others/dimention_reduce_training.md b/tutorials/experts/source_zh_cn/others/dimention_reduce_training.md index 87f478b1d8dab75f2560205906185819b3748e81..bb7a53364d8e8e2ba0c39146ba0ef8d1af402359 100644 --- a/tutorials/experts/source_zh_cn/others/dimention_reduce_training.md +++ b/tutorials/experts/source_zh_cn/others/dimention_reduce_training.md @@ -22,7 +22,7 @@ ### 配置分布式环境变量 -在本地Ascend处理器上进行分布式训练时，需要配置当前多卡环境的组网信息文件，1个8卡环境的json文件配置如下，本样例将该配置文件命名为rank_table_8pcs.json。rank_table可以使用models下面的[hccl_tools.py](https://gitee.com/mindspore/models/blob/master/utils/hccl_tools/hccl_tools.py)生成。 +在本地Ascend处理器上进行分布式训练时，需要配置当前多卡环境的组网信息文件，1个8卡环境的json文件配置如下，本样例将该配置文件命名为rank_table_8pcs.json。rank_table可以使用models下面的[hccl_tools.py](https://gitee.com/mindspore/models/blob/r1.7/utils/hccl_tools/hccl_tools.py)生成。 ```json { @@ -86,7 +86,7 @@ init() ### 加载数据集 -利用MindSpore提供图片加载接口ImageFolderDataset加载ImageNet 2012数据集，同时通过MindSpore提供的数据增强接口对数据集进行处理，此部分代码由models中`resnet`目录下的[dataset.py](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/src/dataset.py)导入。 +利用MindSpore提供图片加载接口ImageFolderDataset加载ImageNet 2012数据集，同时通过MindSpore提供的数据增强接口对数据集进行处理，此部分代码由models中`resnet`目录下的[dataset.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/src/dataset.py)导入。 ```python # define train dataset @@ -110,9 +110,9 @@ init_weight(net=net) 定义模型所需的损失函数loss、optimizer等。 -loss使用CrossEntropySmooth，由ModelZoo中`resnet`目录下的[CrossEntropySmooth.py](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/src/CrossEntropySmooth.py)导入。 +loss使用CrossEntropySmooth，由ModelZoo中`resnet`目录下的[CrossEntropySmooth.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/src/CrossEntropySmooth.py)导入。 -学习率lr的构建代码由models中`resnet`目录下的[lr_generator.py](https://gitee.com/mindspore/models/blob/master/official/cv/resnet/src/lr_generator.py)导入。 +学习率lr的构建代码由models中`resnet`目录下的[lr_generator.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/resnet/src/lr_generator.py)导入。 ```python # define loss diff --git a/tutorials/experts/source_zh_cn/others/gradient_accumulation.md b/tutorials/experts/source_zh_cn/others/gradient_accumulation.md index 8031acc20067a1c8650556f4d5c3551cc1b019f1..7b7e81fbf61b5f0d956fe4a1e78c58f1e4a13784 100644 --- a/tutorials/experts/source_zh_cn/others/gradient_accumulation.md +++ b/tutorials/experts/source_zh_cn/others/gradient_accumulation.md @@ -93,11 +93,11 @@ from models.official.cv.lenet.src.lenet import LeNet5 #### 加载数据集 -利用MindSpore的`dataset`提供的`MnistDataset`接口加载MNIST数据集，此部分代码由models中`lenet`目录下的[dataset.py](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/src/dataset.py)导入。 +利用MindSpore的`dataset`提供的`MnistDataset`接口加载MNIST数据集，此部分代码由models中`lenet`目录下的[dataset.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/src/dataset.py)导入。 #### 定义网络 -这里以LeNet网络为例进行介绍，当然也可以使用其它的网络，如ResNet-50、BERT等, 此部分代码由models中`lenet`目录下的[lenet.py](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/src/lenet.py)导入。 +这里以LeNet网络为例进行介绍，当然也可以使用其它的网络，如ResNet-50、BERT等, 此部分代码由models中`lenet`目录下的[lenet.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/src/lenet.py)导入。 #### 定义训练流程 @@ -289,7 +289,7 @@ if __name__ == "__main__": **验证模型:** -通过ModelZoo中`lenet`目录下的[eval.py](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/eval.py)，使用保存的CheckPoint文件，加载验证数据集，进行验证。 +通过ModelZoo中`lenet`目录下的[eval.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/eval.py)，使用保存的CheckPoint文件，加载验证数据集，进行验证。 ```bash python eval.py --data_path=./MNIST_Data --ckpt_path=./gradient_accumulation.ckpt --device_target=GPU @@ -330,11 +330,11 @@ from models.official.cv.lenet.src.lenet import LeNet5 #### 加载数据集 -利用MindSpore的`dataset`提供的`MnistDataset`接口加载MNIST数据集，此部分代码由models中`lenet`目录下的[dataset.py](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/src/dataset.py)导入。 +利用MindSpore的`dataset`提供的`MnistDataset`接口加载MNIST数据集，此部分代码由models中`lenet`目录下的[dataset.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/src/dataset.py)导入。 #### 定义网络 -这里以LeNet网络为例进行介绍，当然也可以使用其它的网络，如ResNet-50、BERT等, 此部分代码由models中`lenet`目录下的[lenet.py](https://gitee.com/mindspore/models/blob/master/official/cv/lenet/src/lenet.py)导入。 +这里以LeNet网络为例进行介绍，当然也可以使用其它的网络，如ResNet-50、BERT等, 此部分代码由models中`lenet`目录下的[lenet.py](https://gitee.com/mindspore/models/blob/r1.7/official/cv/lenet/src/lenet.py)导入。 #### 定义训练模型 diff --git a/tutorials/experts/source_zh_cn/parallel/pangu_alpha.md b/tutorials/experts/source_zh_cn/parallel/pangu_alpha.md index 40dd613bb1f95340291d37352df034dfc64e1a19..ee93828de4a53c99daa7a8f638897bfe02c69a89 100644 --- a/tutorials/experts/source_zh_cn/parallel/pangu_alpha.md +++ b/tutorials/experts/source_zh_cn/parallel/pangu_alpha.md @@ -6,7 +6,7 @@ 在MindSpore发布的鹏程·盘古模型[1]中，我们看到借助多维度自动混合并行可以实现超大规模Transformer网络的分布式训练。这篇文章将从网络脚本出发，详解模型各个组成部分的切分方式。 -> 完整代码可以参考：https://gitee.com/mindspore/models/tree/master/official/nlp/pangu_alpha +> 完整代码可以参考：https://gitee.com/mindspore/models/tree/r1.7/official/nlp/pangu_alpha 在训练入口脚本train.py中，通过`context.set_auto_parallel_context`接口使能半自动并行模式`SEMI_AUTO_PARALLEL`，表明用户可以通过对算子配置切分策略的方式，借助框架自动完成切分。根据不同网络层运算量和计算方式的特点，选择合适的切分策略是本文关注的重点。此外，通过`enable_parallel_optimizer`和`pipeline_stages`参数可以配置优化器并行和流水线并行方式。