From 0bc804fffb779db8447d4d2b122f8220ded1d0ea Mon Sep 17 00:00:00 2001 From: wangjiangben Date: Mon, 5 Dec 2022 10:23:34 +0800 Subject: [PATCH 01/14] update NPU --> ASCEND --- docs/en/understand_mmcv/ops.md | 116 +++++++++++++++--------------- docs/zh_cn/understand_mmcv/ops.md | 116 +++++++++++++++--------------- 2 files changed, 116 insertions(+), 116 deletions(-) diff --git a/docs/en/understand_mmcv/ops.md b/docs/en/understand_mmcv/ops.md index 822ee15589..e87fda2a76 100644 --- a/docs/en/understand_mmcv/ops.md +++ b/docs/en/understand_mmcv/ops.md @@ -2,61 +2,61 @@ We implement common ops used in detection, segmentation, etc. -| Device | CPU | CUDA | MLU | MPS | NPU | -| ---------------------------- | --- | ---- | --- | --- | --- | -| ActiveRotatedFilter | √ | √ | | | | -| AssignScoreWithK | | √ | | | | -| BallQuery | | √ | | | | -| BBoxOverlaps | | √ | √ | √ | | -| BorderAlign | | √ | | | | -| BoxIouRotated | √ | √ | | | | -| BoxIouQuadri | √ | √ | | | | -| CARAFE | | √ | √ | | | -| ChamferDistance | | √ | | | | -| CrissCrossAttention | | √ | | | | -| ContourExpand | √ | | | | | -| ConvexIoU | | √ | | | | -| CornerPool | | √ | | | | -| Correlation | | √ | | | | -| Deformable Convolution v1/v2 | √ | √ | | | | -| Deformable RoIPool | | √ | √ | | | -| DiffIoURotated | | √ | | | | -| DynamicScatter | | √ | | | | -| FurthestPointSample | | √ | | | | -| FurthestPointSampleWithDist | | √ | | | | -| FusedBiasLeakyrelu | | √ | | | | -| GatherPoints | | √ | | | | -| GroupPoints | | √ | | | | -| Iou3d | | √ | √ | | | -| KNN | | √ | | | | -| MaskedConv | | √ | √ | | √ | -| MergeCells | | √ | | | | -| MinAreaPolygon | | √ | | | | -| ModulatedDeformConv2d | √ | √ | √ | | √ | -| MultiScaleDeformableAttn | | √ | √ | | | -| NMS | √ | √ | √ | | √ | -| NMSRotated | √ | √ | | | | -| NMSQuadri | √ | √ | | | | -| PixelGroup | √ | | | | | -| PointsInBoxes | √ | √ | | | | -| PointsInPolygons | | √ | | | | -| PSAMask | √ | √ | √ | | | -| RotatedFeatureAlign | √ | √ | | | | -| RoIPointPool3d | | √ | √ | | | -| RoIPool | | √ | √ | | | -| RoIAlignRotated | √ | √ | √ | | | -| RiRoIAlignRotated | | √ | | | | -| RoIAlign | √ | √ | √ | | | -| RoIAwarePool3d | | √ | √ | | | -| SAConv2d | | √ | | | | -| SigmoidFocalLoss | | √ | √ | | √ | -| SoftmaxFocalLoss | | √ | | | √ | -| SoftNMS | | √ | | | | -| Sparse Convolution | | √ | | | | -| Synchronized BatchNorm | | √ | | | | -| ThreeInterpolate | | √ | | | | -| ThreeNN | | √ | √ | | | -| TINShift | | √ | √ | | | -| UpFirDn2d | | √ | | | | -| Voxelization | √ | √ | | | | -| PrRoIPool | | √ | | | | +| Device | CPU | CUDA | MLU | MPS | ASCEND | +| ---------------------------- | --- | ---- | --- | --- | ------ | +| ActiveRotatedFilter | √ | √ | | | | +| AssignScoreWithK | | √ | | | | +| BallQuery | | √ | | | | +| BBoxOverlaps | | √ | √ | √ | | +| BorderAlign | | √ | | | | +| BoxIouRotated | √ | √ | | | | +| BoxIouQuadri | √ | √ | | | | +| CARAFE | | √ | √ | | | +| ChamferDistance | | √ | | | | +| CrissCrossAttention | | √ | | | | +| ContourExpand | √ | | | | | +| ConvexIoU | | √ | | | | +| CornerPool | | √ | | | | +| Correlation | | √ | | | | +| Deformable Convolution v1/v2 | √ | √ | | | | +| Deformable RoIPool | | √ | √ | | | +| DiffIoURotated | | √ | | | | +| DynamicScatter | | √ | | | | +| FurthestPointSample | | √ | | | | +| FurthestPointSampleWithDist | | √ | | | | +| FusedBiasLeakyrelu | | √ | | | | +| GatherPoints | | √ | | | | +| GroupPoints | | √ | | | | +| Iou3d | | √ | √ | | | +| KNN | | √ | | | | +| MaskedConv | | √ | √ | | √ | +| MergeCells | | √ | | | | +| MinAreaPolygon | | √ | | | | +| ModulatedDeformConv2d | √ | √ | √ | | √ | +| MultiScaleDeformableAttn | | √ | √ | | | +| NMS | √ | √ | √ | | √ | +| NMSRotated | √ | √ | | | | +| NMSQuadri | √ | √ | | | | +| PixelGroup | √ | | | | | +| PointsInBoxes | √ | √ | | | | +| PointsInPolygons | | √ | | | | +| PSAMask | √ | √ | √ | | | +| RotatedFeatureAlign | √ | √ | | | | +| RoIPointPool3d | | √ | √ | | | +| RoIPool | | √ | √ | | | +| RoIAlignRotated | √ | √ | √ | | | +| RiRoIAlignRotated | | √ | | | | +| RoIAlign | √ | √ | √ | | | +| RoIAwarePool3d | | √ | √ | | | +| SAConv2d | | √ | | | | +| SigmoidFocalLoss | | √ | √ | | √ | +| SoftmaxFocalLoss | | √ | | | √ | +| SoftNMS | | √ | | | | +| Sparse Convolution | | √ | | | | +| Synchronized BatchNorm | | √ | | | | +| ThreeInterpolate | | √ | | | | +| ThreeNN | | √ | √ | | | +| TINShift | | √ | √ | | | +| UpFirDn2d | | √ | | | | +| Voxelization | √ | √ | | | | +| PrRoIPool | | √ | | | | diff --git a/docs/zh_cn/understand_mmcv/ops.md b/docs/zh_cn/understand_mmcv/ops.md index 23d9b6e5fd..6ce4872b94 100644 --- a/docs/zh_cn/understand_mmcv/ops.md +++ b/docs/zh_cn/understand_mmcv/ops.md @@ -2,61 +2,61 @@ MMCV 提供了检测、分割等任务中常用的算子 -| Device | CPU | CUDA | MLU | MPS | NPU | -| ---------------------------- | --- | ---- | --- | --- | --- | -| ActiveRotatedFilter | √ | √ | | | | -| AssignScoreWithK | | √ | | | | -| BallQuery | | √ | | | | -| BBoxOverlaps | | √ | √ | √ | | -| BorderAlign | | √ | | | | -| BoxIouRotated | √ | √ | | | | -| BoxIouQuadri | √ | √ | | | | -| CARAFE | | √ | √ | | | -| ChamferDistance | | √ | | | | -| CrissCrossAttention | | √ | | | | -| ContourExpand | √ | | | | | -| ConvexIoU | | √ | | | | -| CornerPool | | √ | | | | -| Correlation | | √ | | | | -| Deformable Convolution v1/v2 | √ | √ | | | | -| Deformable RoIPool | | √ | √ | | | -| DiffIoURotated | | √ | | | | -| DynamicScatter | | √ | | | | -| FurthestPointSample | | √ | | | | -| FurthestPointSampleWithDist | | √ | | | | -| FusedBiasLeakyrelu | | √ | | | | -| GatherPoints | | √ | | | | -| GroupPoints | | √ | | | | -| Iou3d | | √ | √ | | | -| KNN | | √ | | | | -| MaskedConv | | √ | √ | | √ | -| MergeCells | | √ | | | | -| MinAreaPolygon | | √ | | | | -| ModulatedDeformConv2d | √ | √ | √ | | √ | -| MultiScaleDeformableAttn | | √ | √ | | | -| NMS | √ | √ | √ | | √ | -| NMSRotated | √ | √ | | | | -| NMSQuadri | √ | √ | | | | -| PixelGroup | √ | | | | | -| PointsInBoxes | √ | √ | | | | -| PointsInPolygons | | √ | | | | -| PSAMask | √ | √ | √ | | | -| RotatedFeatureAlign | √ | √ | | | | -| RoIPointPool3d | | √ | √ | | | -| RoIPool | | √ | √ | | | -| RoIAlignRotated | √ | √ | √ | | | -| RiRoIAlignRotated | | √ | | | | -| RoIAlign | √ | √ | √ | | | -| RoIAwarePool3d | | √ | √ | | | -| SAConv2d | | √ | | | | -| SigmoidFocalLoss | | √ | √ | | √ | -| SoftmaxFocalLoss | | √ | | | √ | -| SoftNMS | | √ | | | | -| Sparse Convolution | | √ | | | | -| Synchronized BatchNorm | | √ | | | | -| ThreeInterpolate | | √ | | | | -| ThreeNN | | √ | √ | | | -| TINShift | | √ | √ | | | -| UpFirDn2d | | √ | | | | -| Voxelization | √ | √ | | | | -| PrRoIPool | | √ | | | | +| Device | CPU | CUDA | MLU | MPS | ASCEND | +| ---------------------------- | --- | ---- | --- | --- | ------ | +| ActiveRotatedFilter | √ | √ | | | | +| AssignScoreWithK | | √ | | | | +| BallQuery | | √ | | | | +| BBoxOverlaps | | √ | √ | √ | | +| BorderAlign | | √ | | | | +| BoxIouRotated | √ | √ | | | | +| BoxIouQuadri | √ | √ | | | | +| CARAFE | | √ | √ | | | +| ChamferDistance | | √ | | | | +| CrissCrossAttention | | √ | | | | +| ContourExpand | √ | | | | | +| ConvexIoU | | √ | | | | +| CornerPool | | √ | | | | +| Correlation | | √ | | | | +| Deformable Convolution v1/v2 | √ | √ | | | | +| Deformable RoIPool | | √ | √ | | | +| DiffIoURotated | | √ | | | | +| DynamicScatter | | √ | | | | +| FurthestPointSample | | √ | | | | +| FurthestPointSampleWithDist | | √ | | | | +| FusedBiasLeakyrelu | | √ | | | | +| GatherPoints | | √ | | | | +| GroupPoints | | √ | | | | +| Iou3d | | √ | √ | | | +| KNN | | √ | | | | +| MaskedConv | | √ | √ | | √ | +| MergeCells | | √ | | | | +| MinAreaPolygon | | √ | | | | +| ModulatedDeformConv2d | √ | √ | √ | | √ | +| MultiScaleDeformableAttn | | √ | √ | | | +| NMS | √ | √ | √ | | √ | +| NMSRotated | √ | √ | | | | +| NMSQuadri | √ | √ | | | | +| PixelGroup | √ | | | | | +| PointsInBoxes | √ | √ | | | | +| PointsInPolygons | | √ | | | | +| PSAMask | √ | √ | √ | | | +| RotatedFeatureAlign | √ | √ | | | | +| RoIPointPool3d | | √ | √ | | | +| RoIPool | | √ | √ | | | +| RoIAlignRotated | √ | √ | √ | | | +| RiRoIAlignRotated | | √ | | | | +| RoIAlign | √ | √ | √ | | | +| RoIAwarePool3d | | √ | √ | | | +| SAConv2d | | √ | | | | +| SigmoidFocalLoss | | √ | √ | | √ | +| SoftmaxFocalLoss | | √ | | | √ | +| SoftNMS | | √ | | | | +| Sparse Convolution | | √ | | | | +| Synchronized BatchNorm | | √ | | | | +| ThreeInterpolate | | √ | | | | +| ThreeNN | | √ | √ | | | +| TINShift | | √ | √ | | | +| UpFirDn2d | | √ | | | | +| Voxelization | √ | √ | | | | +| PrRoIPool | | √ | | | | From c3a9c9d071748efc8cbd318261fca5860c5f3e5a Mon Sep 17 00:00:00 2001 From: wangjiangben Date: Mon, 5 Dec 2022 11:49:40 +0800 Subject: [PATCH 02/14] update docs --- docs/en/get_started/build.md | 48 +++++++++++++++++++++++++++++++++ docs/zh_cn/get_started/build.md | 35 ++++++++++++++---------- 2 files changed, 69 insertions(+), 14 deletions(-) diff --git a/docs/en/get_started/build.md b/docs/en/get_started/build.md index c03bea5912..45a9de7123 100644 --- a/docs/en/get_started/build.md +++ b/docs/en/get_started/build.md @@ -327,3 +327,51 @@ Firstly, you need to apply for an IPU cloud machine, see [here](https://www.grap 1. Build MMCV 2. Use pip to install sdk according to [IPU PyTorch document](https://docs.graphcore.ai/projects/poptorch-user-guide/en/latest/installation.html). Also, you need to apply for machine and sdk to Graphcore. + +### Build mmcv-full on ASCEND NPU machine + +- Firstly, see torch_npu for the complete installation tutorial [PyTorch Installation Guide](https://gitee.com/ascend/pytorch/blob/master/docs/en/PyTorch%20Installation%20Guide/PyTorch%20Installation%20Guide.md) + +#### Option 1: Install mmcv-full with pip + +- The NPU compiled version of mmcv-full is already supported when the version of mmcv >= 1.7.0, we can pip install directly + +```bash +pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/npu/torch1.8.0/index.html +``` + +#### Option 2: Build mmcv-full NPU(ASCEND) from Source + +- Pull the source code + +```bash +git pull https://github.com/open-mmlab/mmcv/tree/master +``` + +- Build + +```bash +MMCV_WITH_OPS=1 MAX_JOBS=8 FORCE_NPU=1 python setup.py build_ext +``` + +- Install + +```bash +MMCV_WITH_OPS=1 FORCE_NPU=1 python setup.py develop +``` + +#### Test Case + +```python +import torch +import torch_npu +from mmcv.ops import softmax_focal_loss + +# Init tensor to the NPU +x = torch.randn(3, 10).npu() +y = torch.tensor([1, 5, 3]).npu() +w = torch.ones(10).float().npu() + +output = softmax_focal_loss(x, y, 2.0, 0.25, w, 'none') +print(output) +``` diff --git a/docs/zh_cn/get_started/build.md b/docs/zh_cn/get_started/build.md index 99c6c7f593..a48cb29a93 100644 --- a/docs/zh_cn/get_started/build.md +++ b/docs/zh_cn/get_started/build.md @@ -344,41 +344,48 @@ mmcv-full 有两个版本： ### 在昇腾 NPU 机器编译 mmcv-full -#### 1. 安装 torch_npu - - torch_npu 完整安装教程详见 [PyTorch安装指南](https://gitee.com/ascend/pytorch/blob/master/docs/zh/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97.md#pytorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97) -#### 2. 编译 MMCV NPU +#### 选项 1: 使用pip安装npu编译版本的mmcv-full + +- NPU编译版本的 mmcv-full 在 mmcv >= 1.7.0 时已经支持直接pip安装 + +```bash +pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/npu/torch1.8.0/index.html +``` -拉取 [MMCV 源码](https://github.com/open-mmlab/mmcv/tree/master) 进行编译安装 +#### 选项 2: 使用NPU设备源码编译安装 mmcv-full -编译： +- 拉取 [MMCV 源码](https://github.com/open-mmlab/mmcv/tree/master) + +```bash +git pull https://github.com/open-mmlab/mmcv/tree/master +``` + +- 编译 ```bash MMCV_WITH_OPS=1 MAX_JOBS=8 FORCE_NPU=1 python setup.py build_ext ``` -安装： +- 安装 ```bash MMCV_WITH_OPS=1 FORCE_NPU=1 python setup.py develop ``` -验证： +#### 验证 ```python import torch import torch_npu from mmcv.ops import softmax_focal_loss + +# Init tensor to the NPU x = torch.randn(3, 10).npu() -x.requires_grad = True y = torch.tensor([1, 5, 3]).npu() w = torch.ones(10).float().npu() -output = softmax_focal_loss(x, y, 2.0, 0.25, w, 'none') -``` -使用说明： - -```{note} -在 tensor 末尾加上 .npu() 就可以调用到 npu 算子 +output = softmax_focal_loss(x, y, 2.0, 0.25, w, 'none') +print(output) ``` From 62c6119d575a73e311aaf63f2c00c84062a75b3c Mon Sep 17 00:00:00 2001 From: wangjiangben Date: Mon, 5 Dec 2022 11:54:28 +0800 Subject: [PATCH 03/14] update --- docs/zh_cn/get_started/build.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/docs/zh_cn/get_started/build.md b/docs/zh_cn/get_started/build.md index a48cb29a93..454333d2be 100644 --- a/docs/zh_cn/get_started/build.md +++ b/docs/zh_cn/get_started/build.md @@ -344,9 +344,9 @@ mmcv-full 有两个版本： ### 在昇腾 NPU 机器编译 mmcv-full -- torch_npu 完整安装教程详见 [PyTorch安装指南](https://gitee.com/ascend/pytorch/blob/master/docs/zh/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97.md#pytorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97) +- torch_npu 完整安装教程详见 [PyTorch 安装指南](https://gitee.com/ascend/pytorch/blob/master/docs/zh/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97.md#pytorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97) -#### 选项 1: 使用pip安装npu编译版本的mmcv-full +#### 选项 1: 使用 pip 安装 npu 编译版本的 mmcv-full - NPU编译版本的 mmcv-full 在 mmcv >= 1.7.0 时已经支持直接pip安装 @@ -354,7 +354,7 @@ mmcv-full 有两个版本： pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/npu/torch1.8.0/index.html ``` -#### 选项 2: 使用NPU设备源码编译安装 mmcv-full +#### 选项 2: 使用 NPU 设备源码编译安装 mmcv-full - 拉取 [MMCV 源码](https://github.com/open-mmlab/mmcv/tree/master) From 5f38cfd7e5a1f0adbf0866d0caa4a2999af8f2ce Mon Sep 17 00:00:00 2001 From: wangjiangben Date: Mon, 5 Dec 2022 14:12:51 +0800 Subject: [PATCH 04/14] update docs --- docs/en/get_started/build.md | 4 ++-- docs/en/understand_mmcv/ops.md | 2 +- docs/zh_cn/understand_mmcv/ops.md | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/docs/en/get_started/build.md b/docs/en/get_started/build.md index 45a9de7123..a102e2e28f 100644 --- a/docs/en/get_started/build.md +++ b/docs/en/get_started/build.md @@ -328,7 +328,7 @@ Firstly, you need to apply for an IPU cloud machine, see [here](https://www.grap 2. Use pip to install sdk according to [IPU PyTorch document](https://docs.graphcore.ai/projects/poptorch-user-guide/en/latest/installation.html). Also, you need to apply for machine and sdk to Graphcore. -### Build mmcv-full on ASCEND NPU machine +### Build mmcv-full on Ascend NPU machine - Firstly, see torch_npu for the complete installation tutorial [PyTorch Installation Guide](https://gitee.com/ascend/pytorch/blob/master/docs/en/PyTorch%20Installation%20Guide/PyTorch%20Installation%20Guide.md) @@ -340,7 +340,7 @@ Firstly, you need to apply for an IPU cloud machine, see [here](https://www.grap pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/npu/torch1.8.0/index.html ``` -#### Option 2: Build mmcv-full NPU(ASCEND) from Source +#### Option 2: Build mmcv-full NPU(Ascend) from Source - Pull the source code diff --git a/docs/en/understand_mmcv/ops.md b/docs/en/understand_mmcv/ops.md index e87fda2a76..058309e25f 100644 --- a/docs/en/understand_mmcv/ops.md +++ b/docs/en/understand_mmcv/ops.md @@ -2,7 +2,7 @@ We implement common ops used in detection, segmentation, etc. -| Device | CPU | CUDA | MLU | MPS | ASCEND | +| Device | CPU | CUDA | MLU | MPS | Ascend | | ---------------------------- | --- | ---- | --- | --- | ------ | | ActiveRotatedFilter | √ | √ | | | | | AssignScoreWithK | | √ | | | | diff --git a/docs/zh_cn/understand_mmcv/ops.md b/docs/zh_cn/understand_mmcv/ops.md index 6ce4872b94..51cc3eec4b 100644 --- a/docs/zh_cn/understand_mmcv/ops.md +++ b/docs/zh_cn/understand_mmcv/ops.md @@ -2,7 +2,7 @@ MMCV 提供了检测、分割等任务中常用的算子 -| Device | CPU | CUDA | MLU | MPS | ASCEND | +| Device | CPU | CUDA | MLU | MPS | Ascend | | ---------------------------- | --- | ---- | --- | --- | ------ | | ActiveRotatedFilter | √ | √ | | | | | AssignScoreWithK | | √ | | | | From f808320726d51cfed48766132ca68102d71a7e02 Mon Sep 17 00:00:00 2001 From: wangjiangben-hw <111729245+wangjiangben-hw@users.noreply.github.com> Date: Tue, 6 Dec 2022 19:25:04 +0800 Subject: [PATCH 05/14] Update docs/en/get_started/build.md Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> --- docs/en/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/get_started/build.md b/docs/en/get_started/build.md index a102e2e28f..7dd892e0eb 100644 --- a/docs/en/get_started/build.md +++ b/docs/en/get_started/build.md @@ -337,7 +337,7 @@ Firstly, you need to apply for an IPU cloud machine, see [here](https://www.grap - The NPU compiled version of mmcv-full is already supported when the version of mmcv >= 1.7.0, we can pip install directly ```bash -pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/npu/torch1.8.0/index.html +pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/ascend/torch1.8.0/index.html ``` #### Option 2: Build mmcv-full NPU(Ascend) from Source From 94c1d1ef69fe203dca70eda6ece039a35c39cb82 Mon Sep 17 00:00:00 2001 From: wangjiangben-hw <111729245+wangjiangben-hw@users.noreply.github.com> Date: Tue, 6 Dec 2022 19:25:11 +0800 Subject: [PATCH 06/14] Update docs/zh_cn/get_started/build.md Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> --- docs/zh_cn/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/zh_cn/get_started/build.md b/docs/zh_cn/get_started/build.md index 454333d2be..45bf38ac21 100644 --- a/docs/zh_cn/get_started/build.md +++ b/docs/zh_cn/get_started/build.md @@ -351,7 +351,7 @@ mmcv-full 有两个版本： - NPU编译版本的 mmcv-full 在 mmcv >= 1.7.0 时已经支持直接pip安装 ```bash -pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/npu/torch1.8.0/index.html +pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/ascend/torch1.8.0/index.html ``` #### 选项 2: 使用 NPU 设备源码编译安装 mmcv-full From 6adc79ba60b069eff8c4b8974d797a57e1165e81 Mon Sep 17 00:00:00 2001 From: wangjiangben-hw <111729245+wangjiangben-hw@users.noreply.github.com> Date: Wed, 7 Dec 2022 14:52:07 +0800 Subject: [PATCH 07/14] Update docs/zh_cn/get_started/build.md Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> --- docs/zh_cn/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/zh_cn/get_started/build.md b/docs/zh_cn/get_started/build.md index 45bf38ac21..e6240a7b46 100644 --- a/docs/zh_cn/get_started/build.md +++ b/docs/zh_cn/get_started/build.md @@ -344,7 +344,7 @@ mmcv-full 有两个版本： ### 在昇腾 NPU 机器编译 mmcv-full -- torch_npu 完整安装教程详见 [PyTorch 安装指南](https://gitee.com/ascend/pytorch/blob/master/docs/zh/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97.md#pytorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97) +在编译 mmcv-full 前，需要安装 torch_npu，完整安装教程详见 [PyTorch 安装指南](https://gitee.com/ascend/pytorch/blob/master/docs/zh/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97.md#pytorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97) #### 选项 1: 使用 pip 安装 npu 编译版本的 mmcv-full From ee5e2828e00e385a4041d09d8b84ae52e8338692 Mon Sep 17 00:00:00 2001 From: wangjiangben-hw <111729245+wangjiangben-hw@users.noreply.github.com> Date: Wed, 7 Dec 2022 14:52:15 +0800 Subject: [PATCH 08/14] Update docs/zh_cn/get_started/build.md Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> --- docs/zh_cn/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/zh_cn/get_started/build.md b/docs/zh_cn/get_started/build.md index e6240a7b46..80fb0303d6 100644 --- a/docs/zh_cn/get_started/build.md +++ b/docs/zh_cn/get_started/build.md @@ -346,7 +346,7 @@ mmcv-full 有两个版本： 在编译 mmcv-full 前，需要安装 torch_npu，完整安装教程详见 [PyTorch 安装指南](https://gitee.com/ascend/pytorch/blob/master/docs/zh/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97/PyTorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97.md#pytorch%E5%AE%89%E8%A3%85%E6%8C%87%E5%8D%97) -#### 选项 1: 使用 pip 安装 npu 编译版本的 mmcv-full +#### 选项 1: 使用 pip 安装 Ascend 编译版本的 mmcv-full - NPU编译版本的 mmcv-full 在 mmcv >= 1.7.0 时已经支持直接pip安装 From e122f39c0f44db901e210b8db199371bba04483c Mon Sep 17 00:00:00 2001 From: wangjiangben-hw <111729245+wangjiangben-hw@users.noreply.github.com> Date: Wed, 7 Dec 2022 14:52:26 +0800 Subject: [PATCH 09/14] Update docs/zh_cn/get_started/build.md Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> --- docs/zh_cn/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/zh_cn/get_started/build.md b/docs/zh_cn/get_started/build.md index 80fb0303d6..221943bea4 100644 --- a/docs/zh_cn/get_started/build.md +++ b/docs/zh_cn/get_started/build.md @@ -348,7 +348,7 @@ mmcv-full 有两个版本： #### 选项 1: 使用 pip 安装 Ascend 编译版本的 mmcv-full -- NPU编译版本的 mmcv-full 在 mmcv >= 1.7.0 时已经支持直接pip安装 +- Ascend 编译版本的 mmcv-full 在 mmcv >= 1.7.0 时已经支持直接 pip 安装 ```bash pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/ascend/torch1.8.0/index.html From 0faf71738a4d74671fd6976115177fb0fa3ea14b Mon Sep 17 00:00:00 2001 From: wangjiangben-hw <111729245+wangjiangben-hw@users.noreply.github.com> Date: Wed, 7 Dec 2022 14:52:38 +0800 Subject: [PATCH 10/14] Update docs/en/get_started/build.md Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> --- docs/en/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/get_started/build.md b/docs/en/get_started/build.md index 7dd892e0eb..355fefb2ca 100644 --- a/docs/en/get_started/build.md +++ b/docs/en/get_started/build.md @@ -340,7 +340,7 @@ Firstly, you need to apply for an IPU cloud machine, see [here](https://www.grap pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/ascend/torch1.8.0/index.html ``` -#### Option 2: Build mmcv-full NPU(Ascend) from Source +#### Option 2: Build mmcv-full NPU (Ascend) from Source - Pull the source code From 79ed0c2da5b6bd65ad08813ce106d719673931a4 Mon Sep 17 00:00:00 2001 From: wangjiangben-hw <111729245+wangjiangben-hw@users.noreply.github.com> Date: Wed, 7 Dec 2022 14:52:49 +0800 Subject: [PATCH 11/14] Update docs/en/get_started/build.md Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> --- docs/en/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/get_started/build.md b/docs/en/get_started/build.md index 355fefb2ca..c23432e472 100644 --- a/docs/en/get_started/build.md +++ b/docs/en/get_started/build.md @@ -334,7 +334,7 @@ Firstly, you need to apply for an IPU cloud machine, see [here](https://www.grap #### Option 1: Install mmcv-full with pip -- The NPU compiled version of mmcv-full is already supported when the version of mmcv >= 1.7.0, we can pip install directly +The Ascend compiled version of mmcv-full is already supported when the version of mmcv >= 1.7.0, we can pip install directly ```bash pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/ascend/torch1.8.0/index.html From 8ca9d2716099568b1185cdaf380a9cd30a31e1e8 Mon Sep 17 00:00:00 2001 From: wangjiangben-hw <111729245+wangjiangben-hw@users.noreply.github.com> Date: Wed, 7 Dec 2022 14:53:03 +0800 Subject: [PATCH 12/14] Update docs/en/get_started/build.md Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> --- docs/en/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/get_started/build.md b/docs/en/get_started/build.md index c23432e472..4e255b6c7f 100644 --- a/docs/en/get_started/build.md +++ b/docs/en/get_started/build.md @@ -330,7 +330,7 @@ Firstly, you need to apply for an IPU cloud machine, see [here](https://www.grap ### Build mmcv-full on Ascend NPU machine -- Firstly, see torch_npu for the complete installation tutorial [PyTorch Installation Guide](https://gitee.com/ascend/pytorch/blob/master/docs/en/PyTorch%20Installation%20Guide/PyTorch%20Installation%20Guide.md) +- Before building mmcv-full, `torch_npu` should be installed. See the complete installation tutorial [PyTorch Installation Guide](https://gitee.com/ascend/pytorch/blob/master/docs/en/PyTorch%20Installation%20Guide/PyTorch%20Installation%20Guide.md) #### Option 1: Install mmcv-full with pip From 196df221cf3d24c1103a2dfbd8a5ecae6ec893fb Mon Sep 17 00:00:00 2001 From: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> Date: Wed, 7 Dec 2022 14:56:42 +0800 Subject: [PATCH 13/14] Update docs/en/get_started/build.md --- docs/en/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/get_started/build.md b/docs/en/get_started/build.md index 4e255b6c7f..091bde838b 100644 --- a/docs/en/get_started/build.md +++ b/docs/en/get_started/build.md @@ -330,7 +330,7 @@ Firstly, you need to apply for an IPU cloud machine, see [here](https://www.grap ### Build mmcv-full on Ascend NPU machine -- Before building mmcv-full, `torch_npu` should be installed. See the complete installation tutorial [PyTorch Installation Guide](https://gitee.com/ascend/pytorch/blob/master/docs/en/PyTorch%20Installation%20Guide/PyTorch%20Installation%20Guide.md) +Before building mmcv-full, `torch_npu` should be installed. See the complete installation tutorial [PyTorch Installation Guide](https://gitee.com/ascend/pytorch/blob/master/docs/en/PyTorch%20Installation%20Guide/PyTorch%20Installation%20Guide.md) #### Option 1: Install mmcv-full with pip From 046dcdacfd3bc9b842930e47ee6e5eb80e9a01e0 Mon Sep 17 00:00:00 2001 From: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com> Date: Wed, 7 Dec 2022 14:57:02 +0800 Subject: [PATCH 14/14] Update docs/zh_cn/get_started/build.md --- docs/zh_cn/get_started/build.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/zh_cn/get_started/build.md b/docs/zh_cn/get_started/build.md index 221943bea4..2098faa71c 100644 --- a/docs/zh_cn/get_started/build.md +++ b/docs/zh_cn/get_started/build.md @@ -348,7 +348,7 @@ mmcv-full 有两个版本： #### 选项 1: 使用 pip 安装 Ascend 编译版本的 mmcv-full -- Ascend 编译版本的 mmcv-full 在 mmcv >= 1.7.0 时已经支持直接 pip 安装 +Ascend 编译版本的 mmcv-full 在 mmcv >= 1.7.0 时已经支持直接 pip 安装 ```bash pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/ascend/torch1.8.0/index.html