Skip to content

Commit

Permalink
[Doc]: onnxruntime (open-mmlab#131)
Browse files Browse the repository at this point in the history 
* add ort doc

* update

* update

* update
  • Loading branch information
@RunningLeon
RunningLeon authored Oct 19, 2021
1 parent 4c1f62f commit a7111ed
Show file tree
Hide file tree
Showing 3 changed files with 219 additions and 4 deletions.
10 changes: 7 additions & 3 deletions docs/backends/ncnn.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@ You should ensure your gcc satisfies `gcc >= 6`.
cmake -DNCNN_VULKAN=ON -DNCNN_SYSTEM_GLSLANG=ON -DNCNN_BUILD_EXAMPLES=ON -DNCNN_PYTHON=ON -DNCNN_BUILD_TOOLS=ON -DNCNN_BUILD_BENCHMARK=ON -DNCNN_BUILD_TESTS=ON ..
make install
```

- Install pyncnn module
```bash
cd ncnn/python
Expand All @@ -42,24 +43,27 @@ cmake -DBUILD_NCNN_OPS=ON ..
make -j$(nproc)
```

If you haven't installed NCNN in default path, please add `-DNCNN_DIR` flag in cmake.
If you haven't installed NCNN in the default path, please add `-DNCNN_DIR` flag in cmake.
```bash
cmake -DBUILD_NCNN_OPS=ON -DNCNN_DIR=${NCNN_DIR} ..
make -j$(nproc)
```
### Convert model
- This follows the tutorial on [How to convert model](tutorials/how_to_convert_model.md).
- This follows the tutorial on [How to convert model](../tutorials/how_to_convert_model.md).
- The converted model has two files: `.param` and `.bin`, as model structure file and weight file respectively.
### FAQs
1. When running ncnn models for inference with custom ops, it fails and shows the error message like:
```
```bash
TypeError: register mm custom layers(): incompatible function arguments. The following argument types are supported:
1.(ar0: ncnn:Net) -> int
Invoked with: <ncnn.ncnn.Net object at 0x7f7fc4038bb0>
```
This is because of the failure to bind ncnn C++ library to pyncnn. You should build pyncnn from C++ ncnn source code, but not by `pip install`
76 changes: 76 additions & 0 deletions docs/backends/onnxruntime.md
View file
Original file line number Diff line number Diff line change
@@ -1,3 +1,79 @@
## ONNX Runtime Support

### Introduction of ONNX Runtime

**ONNX Runtime** is a cross-platform inferencing and training accelerator compatible with many popular ML/DNN frameworks. Check its [github](https://github.com/microsoft/onnxruntime) for more information.

### Installation

*Please note that only **onnxruntime>=1.8.1** of CPU version on Linux platform is supported by now.*

- Install ONNX Runtime python package

```bash
pip install onnxruntime==1.8.1
```

### Build custom ops

#### Prerequisite

- Download `onnxruntime-linux` from ONNX Runtime [releases](https://github.com/microsoft/onnxruntime/releases/tag/v1.8.1), extract it, expose `ONNXRUNTIME_DIR` and finally add the lib path to `LD_LIBRARY_PATH` as below:

```bash
wget https://github.com/microsoft/onnxruntime/releases/download/v1.8.1/onnxruntime-linux-x64-1.8.1.tgz

tar -zxvf onnxruntime-linux-x64-1.8.1.tgz
cd onnxruntime-linux-x64-1.8.1
export ONNXRUNTIME_DIR=$(pwd)
export LD_LIBRARY_PATH=$ONNXRUNTIME_DIR/lib:$LD_LIBRARY_PATH
```

#### Build on Linux

```bash
cd ${MMDEPLOY_DIR} # To MMDeploy root directory
mkdir build
cd build
cmake -DBUILD_ONNXRUNTIME_OPS=ON -DONNXRUNTIME_DIR=${ONNXRUNTIME_DIR} ..
make -j10
```

### How to convert a model

- You could follow the instructions of tutorial [How to convert model](../tutorials/how_to_convert_model.md)

### List of supported custom ops

| Operator | CPU | GPU | MMDeploy Releases |
| :----------------------------------------------------: | :---: | :---: | :-----------: |
| [RoIAlign](../ops/onnxruntime.md#roialign) | Y | N | master |
| [grid_sampler](../ops/onnxruntime.md#grid_sampler) | Y | N | master |
| [MMCVModulatedDeformConv2d](../ops/onnxruntime.md#mmcvmodulateddeformconv2d) | Y | N | master |

### How to add a new custom op

#### Reminder

- The custom operator is not included in [supported operator list](https://github.com/microsoft/onnxruntime/blob/master/docs/OperatorKernels.md) in ONNX Runtime.
- The custom operator should be able to be exported to ONNX.

#### Main procedures

Take custom operator `roi_align` for example.

1. Create a `roi_align` directory in ONNX Runtime source directory `backend_ops/onnxruntime/`
2. Add header and source file into `roi_align` directory `backend_ops/onnxruntime/roi_align/`
3. Add unit test into `tests/test_ops/test_ops.py`
Check [here](../../tests/test_ops/test_ops.py) for examples.

**Finally, welcome to send us PR of adding custom operators for ONNX Runtime in MMDeploy.** :nerd_face:

### FAQs

- None

### References

- [How to export Pytorch model with custom op to ONNX and run it in ONNX Runtime](https://github.com/onnx/tutorials/blob/master/PyTorchCustomOperator/README.md)
- [How to add a custom operator/kernel in ONNX Runtime](https://github.com/microsoft/onnxruntime/blob/master/docs/AddingCustomOp.md)
137 changes: 136 additions & 1 deletion docs/ops/onnxruntime.md
View file
Original file line number Diff line number Diff line change
@@ -1,3 +1,138 @@
## ONNX Runtime Ops

### Installation
<!-- TOC -->

- [ONNX Runtime Ops](#onnx-runtime-ops)
- [RoIAlign](#roialign)
- [Description](#description)
- [Parameters](#parameters)
- [Inputs](#inputs)
- [Outputs](#outputs)
- [Type Constraints](#type-constraints)
- [grid_sampler](#grid_sampler)
- [Description](#description-1)
- [Parameters](#parameters-1)
- [Inputs](#inputs-1)
- [Outputs](#outputs-1)
- [Type Constraints](#type-constraints-1)
- [MMCVModulatedDeformConv2d](#mmcvmodulateddeformconv2d)
- [Description](#description-2)
- [Parameters](#parameters-2)
- [Inputs](#inputs-2)
- [Outputs](#outputs-2)
- [Type Constraints](#type-constraints-2)

<!-- TOC -->

### RoIAlign

#### Description

Perform RoIAlign on output feature, used in bbox_head of most two-stage detectors.

#### Parameters

| Type | Parameter | Description |
| ------- | ---------------- | ------------------------------------------------------------------------------------------------------------- |
| `int` | `output_height` | height of output roi |
| `int` | `output_width` | width of output roi |
| `float` | `spatial_scale` | used to scale the input boxes |
| `int` | `sampling_ratio` | number of input samples to take for each output sample. `0` means to take samples densely for current models. |
| `str` | `mode` | pooling mode in each bin. `avg` or `max` |
| `int` | `aligned` | If `aligned=0`, use the legacy implementation in MMDetection. Else, align the results more perfectly. |

#### Inputs

<dl>
<dt><tt>input</tt>: T</dt>
<dd>Input feature map; 4D tensor of shape (N, C, H, W), where N is the batch size, C is the numbers of channels, H and W are the height and width of the data.</dd>
<dt><tt>rois</tt>: T</dt>
<dd>RoIs (Regions of Interest) to pool over; 2-D tensor of shape (num_rois, 5) given as [[batch_index, x1, y1, x2, y2], ...]. The RoIs' coordinates are the coordinate system of input.</dd>
</dl>

#### Outputs

<dl>
<dt><tt>feat</tt>: T</dt>
<dd>RoI pooled output, 4-D tensor of shape (num_rois, C, output_height, output_width). The r-th batch element feat[r-1] is a pooled feature map corresponding to the r-th RoI RoIs[r-1].<dd>
</dl>

#### Type Constraints

- T:tensor(float32)

### grid_sampler

#### Description

Perform sample from `input` with pixel locations from `grid`.

#### Parameters

| Type | Parameter | Description |
| ----- | -------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `int` | `interpolation_mode` | Interpolation mode to calculate output values. (0: `bilinear` , 1: `nearest`) |
| `int` | `padding_mode` | Padding mode for outside grid values. (0: `zeros`, 1: `border`, 2: `reflection`) |
| `int` | `align_corners` | If `align_corners=1`, the extrema (`-1` and `1`) are considered as referring to the center points of the input's corner pixels. If `align_corners=0`, they are instead considered as referring to the corner points of the input's corner pixels, making the sampling more resolution agnostic. |

#### Inputs

<dl>
<dt><tt>input</tt>: T</dt>
<dd>Input feature; 4-D tensor of shape (N, C, inH, inW), where N is the batch size, C is the numbers of channels, inH and inW are the height and width of the data.</dd>
<dt><tt>grid</tt>: T</dt>
<dd>Input offset; 4-D tensor of shape (N, outH, outW, 2), where outH and outW are the height and width of offset and output. </dd>
</dl>

#### Outputs

<dl>
<dt><tt>output</tt>: T</dt>
<dd>Output feature; 4-D tensor of shape (N, C, outH, outW).</dd>
</dl>

#### Type Constraints

- T:tensor(float32, Linear)

### MMCVModulatedDeformConv2d

#### Description

Perform Modulated Deformable Convolution on input feature, read [Deformable ConvNets v2: More Deformable, Better Results](https://arxiv.org/abs/1811.11168?from=timeline) for detail.

#### Parameters

| Type | Parameter | Description |
| -------------- | ------------------- | ------------------------------------------------------------------------------------- |
| `list of ints` | `stride` | The stride of the convolving kernel. (sH, sW) |
| `list of ints` | `padding` | Paddings on both sides of the input. (padH, padW) |
| `list of ints` | `dilation` | The spacing between kernel elements. (dH, dW) |
| `int` | `deformable_groups` | Groups of deformable offset. |
| `int` | `groups` | Split input into groups. `input_channel` should be divisible by the number of groups. |

#### Inputs

<dl>
<dt><tt>inputs[0]</tt>: T</dt>
<dd>Input feature; 4-D tensor of shape (N, C, inH, inW), where N is the batch size, C is the number of channels, inH and inW are the height and width of the data.</dd>
<dt><tt>inputs[1]</tt>: T</dt>
<dd>Input offset; 4-D tensor of shape (N, deformable_group* 2* kH* kW, outH, outW), where kH and kW are the height and width of weight, outH and outW are the height and width of offset and output.</dd>
<dt><tt>inputs[2]</tt>: T</dt>
<dd>Input mask; 4-D tensor of shape (N, deformable_group* kH* kW, outH, outW), where kH and kW are the height and width of weight, outH and outW are the height and width of offset and output.</dd>
<dt><tt>inputs[3]</tt>: T</dt>
<dd>Input weight; 4-D tensor of shape (output_channel, input_channel, kH, kW).</dd>
<dt><tt>inputs[4]</tt>: T, optional</dt>
<dd>Input bias; 1-D tensor of shape (output_channel).</dd>
</dl>

#### Outputs

<dl>
<dt><tt>outputs[0]</tt>: T</dt>
<dd>Output feature; 4-D tensor of shape (N, output_channel, outH, outW).</dd>
</dl>

#### Type Constraints

- T:tensor(float32, Linear)

0 comments on commit a7111ed

Please sign in to comment.