wulin/rknn-toolkit2
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[update doc] Fix error

Browse Source
This commit is contained in:
Jie 2021-08-30 21:11:09 +08:00
parent 2dc12422cd
commit f45b67e613
4 changed files with 11 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
README.md
View File

@ -14,8 +14,7 @@ RKNN-Toolkit2 is a software development kit for users to perform model conversio
871ca9bbcb96dcf89991e9a6888b3a48 rknn_toolkit2-1.1.0_5c458c6-cp36-cp36m-linux_x86_64.whl 871ca9bbcb96dcf89991e9a6888b3a48 rknn_toolkit2-1.1.0_5c458c6-cp36-cp36m-linux_x86_64.whl
``` ```
# Feedback and Community Suport # Feedback and Community Suport
Three ways are followed: Two ways are followed:
- [Redmine](https://redmine.rock-chips.com/) (**Feedback recommended**)
- [Issues](https://github.com/rockchip-linux/rknn-toolkit2/issues) (**Feedback recommended**) - [Issues](https://github.com/rockchip-linux/rknn-toolkit2/issues) (**Feedback recommended**)
- QQ Group Chat: 1025468710 **welcome for developer** - QQ Group Chat: 1025468710 **welcome for developer**
<p align="center"><img width="200" height="200" src="https://github.com/rockchip-linux/rknn-toolkit2/blob/master/QQGroupQRCode.png"/></p> <p align="center"><img width="200" height="200" src="https://github.com/rockchip-linux/rknn-toolkit2/blob/master/QQGroupQRCode.png"/></p>

18
doc/RKNNToolKit2_OP_Support-1.1.0.md
View File

@ -1,6 +1,6 @@
# RKNNToolkit2 OPs Support # RKNNToolkit2 OPs Support
Base on RKNN Toolkit2 Version 0.6.0 Based on RKNN Toolkit2 Version 1.1.0
## Explanation of terms: ## Explanation of terms:
@ -34,7 +34,7 @@ Broadcast rule:
Caffe protocols RKNN Toolkit2 uses only based on the officially modified protocol of berkeley. Caffe protocols RKNN Toolkit2 uses only based on the officially modified protocol of berkeley.
The protocol based on the official revision of berkeley comes from [berkeley caffe](https://github.com/BVLC/caffe/tree/master/src/caffe/proto 'Berkeley Caffe'), commit hash is 21d0608. On this basis RKNN Toolkit2 have added some OPs. The protocol based on the official revision of berkeley comes from [berkeley caffe](https://github.com/BVLC/caffe/tree/master/src/caffe/proto 'Berkeley Caffe'), commit hash is 21d0608. On this basis RKNN Toolkit2 have added some OPs.
Based on this protocol, the list of Caffe OPs supported by RKNN Toolkit2 Version 0.6.0 is as follows: Based on this protocol, the list of Caffe OPs supported by RKNN Toolkit2 Version 1.1.0 is as follows:
| **Operators** | **Remarks** | | **Operators** | **Remarks** |
| -------------------- | ----------- | | -------------------- | ----------- |
@ -77,8 +77,8 @@ Based on this protocol, the list of Caffe OPs supported by RKNN Toolkit2 Version
## ONNX OPs supported by RKNN Toolkit2 ## ONNX OPs supported by RKNN Toolkit2
The ONNX version supported by RKNN Toolkit2 is 0.6.0. According to [ONNX official instructions](https://github.com/microsoft/onnxruntime/blob/master/docs/Versioning.md 'ONNX Version Description'), the corresponding ONNX opset version is 12. The ONNX version supported by RKNN Toolkit2 is 1.1.0. According to [ONNX official instructions](https://github.com/microsoft/onnxruntime/blob/master/docs/Versioning.md 'ONNX Version Description'), the corresponding ONNX opset version is 12.
The list of ONNX OPs supported by RKNN Toolkit2 Version 0.6.0 is as follows: The list of ONNX OPs supported by RKNN Toolkit2 Version 1.1.0 is as follows:
| **Operators** | **Remarks** | | **Operators** | **Remarks** |
| ----------------------------- | ----------- | | ----------------------------- | ----------- |
@ -119,8 +119,8 @@ The list of ONNX OPs supported by RKNN Toolkit2 Version 0.6.0 is as follows:
| ReduceSum |output dims <= 4| | ReduceSum |output dims <= 4|
| Relu |channel: [1, 8192]<br />height: [1, 8192]<br />width: [1, 8176]| | Relu |channel: [1, 8192]<br />height: [1, 8192]<br />width: [1, 8176]|
| Reshape || | Reshape ||
| Resize |bilinear（不支持tf_crop_and_resize）; nearest2d| | Resize |bilinear（not support tf_crop_and_resize）; nearest2d|
| ReverseSequence | | ReverseSequence ||
| RoiAlign |pool type: average| | RoiAlign |pool type: average|
| Slice || | Slice ||
| Softmax || | Softmax ||
@ -130,12 +130,12 @@ The list of ONNX OPs supported by RKNN Toolkit2 Version 0.6.0 is as follows:
| Tanh |channel: [1, 8192]<br />height: [1, 8192]<br />width: [1, 8176]| | Tanh |channel: [1, 8192]<br />height: [1, 8192]<br />width: [1, 8176]|
| Tile || | Tile ||
| Transpose || | Transpose ||
| Upsample (resize) || | | Upsample (resize) ||
## Pytorch OPs supported by RKNN Toolkit2 ## Pytorch OPs supported by RKNN Toolkit2
The Pytorch version supported by RKNN Toolkit2 is 1.6.0, models generated by other versions may not support. The Pytorch version supported by RKNN Toolkit2 is 1.6.0, models generated by other versions may not support.
The list of Pytorch OPs supported by RKNN Toolkit2 Version 0.6.0 is as follows: The list of Pytorch OPs supported by RKNN Toolkit2 Version 1.1.0 is as follows:
| **Operators** | **Remarks** | | **Operators** | **Remarks** |
| ------------------------- | ----------- | | ------------------------- | ----------- |
@ -242,7 +242,7 @@ The list of TensorFlow Lite OPs supported by RKNN Toolkit2 is as follows:
|TRANSPOSE|| |TRANSPOSE||
## Darknet OPs supported by RKNN Toolkit2 ## Darknet OPs supported by RKNN Toolkit2
The list of Darknet OPs supported by RKNN Toolkit2 Version 0.6.0 is as follows: The list of Darknet OPs supported by RKNN Toolkit2 Version 1.1.0 is as follows:
| **Operators** | **Remarks** | | **Operators** | **Remarks** |
|---| ----------- | |---| ----------- |

2
doc/changelog-1.1.0.txt
View File

@ -1,4 +1,4 @@
2021-8-12 2021-8-20
版本：v1.1.0 版本：v1.1.0
更新内容: 更新内容:
1. 新功能: LSTM，GRU的支持；增加了accuracy_analysis对比项目；增加了一些op支持：caffe hardswish；onnx gather,reduceMax等op；更新了更全面的图优化规则。 1. 新功能: LSTM，GRU的支持；增加了accuracy_analysis对比项目；增加了一些op支持：caffe hardswish；onnx gather,reduceMax等op；更新了更全面的图优化规则。

BIN
packages/rknn_toolkit2-1.1.0_5c458c6-cp36-cp36m-linux_x86_64.whl
View File

Binary file not shown.