beta 0.2.0.7

- move docs to http://www.yuque.com/mnn
- fix bugs for CPU ops TopKV2 and quantized convolution
- add enqueue map buffer error handle for OpenCL
- add nullptr protection for extra tensor desc
- add failure protection for memory acquirement
- fix slice shape calculation
- refactor binary shape calculation

README.md

@@ -39,25 +39,23 @@ Converter consists of Frontends and Graph Optimize. The former is responsible fo
 Interpreter consists of Engine and Backends. The former is responsible for the loading of the model and the scheduling of the calculation graph; the latter includes the memory allocation and the Op implementation under each computing device. In Engine and Backends, MNN applies a variety of optimization schemes, including applying Winograd algorithm in convolution and deconvolution, applying Strassen algorithm in matrix multiplication, low-precision calculation, Neon optimization, hand-written assembly, multi-thread optimization, memory reuse, heterogeneous computing, etc.
 
 ## Quick start
-- [Install](doc/Install_EN.md)
-- [Tutorial](doc/Tutorial_EN.md)
+- [Usage](https://www.yuque.com/mnn/en/usage)
+- [Supported Operators](https://www.yuque.com/mnn/en/ops)
 - [API](doc/API/API_index.html)
-- [Demo](demo)
-- [Model conversion](tools/converter/README.md)
-- [Test tools](doc/Tools_EN.md)
-- [Op list](doc/OpList.md)
-- [Contributing](doc/Contributing_EN.md)
-- [Model Compress](tools/quantization/README_CN.md)
+- [Demo](https://www.yuque.com/mnn/en/demo_zoo)
 
-## Benchmark
-- [Benchmark](doc/Benchmark_EN.md)
+## Tools
+- [Test tools](https://www.yuque.com/mnn/en/tool_test)
+- [Benchmark](https://www.yuque.com/mnn/en/tool_benchmark)
+- [Model compress](https://www.yuque.com/mnn/en/tool_quantize)
 
-## How to customize
-- [Add custom op](doc/AddOp_EN.md)
-- [Add custom backend](doc/AddBackend_EN.md)
+## Customizing
+- [Customizing backend](https://www.yuque.com/mnn/en/customize_backend)
+- [Customizing operator](https://www.yuque.com/mnn/en/customize_op)
+- [Contributing](https://www.yuque.com/mnn/en/contribute)
 
 ## Feedbacks
-- [FAQ](doc/FAQ.md)
+- [FAQ](https://www.yuque.com/mnn/en/faq)
 
 Scan QR code to join DingDing discussion group.
 

README_CN.md

@@ -39,25 +39,23 @@ Converter由Frontends和Graph Optimize构成。前者负责支持不同的训练
 Interpreter由Engine和Backends构成。前者负责模型的加载、计算图的调度；后者包含各计算设备下的内存分配、Op实现。在Engine和Backends中，MNN应用了多种优化方案，包括在卷积和反卷积中应用Winograd算法、在矩阵乘法中应用Strassen算法、低精度计算、Neon优化、手写汇编、多线程优化、内存复用、异构计算等。
 
 ## 开始使用
-- [编译与安装](doc/Install_CN.md)
-- [使用教程](doc/Tutorial_CN.md)
+- [使用方法](https://www.yuque.com/mnn/cn/usage)
+- [算子列表](https://www.yuque.com/mnn/en/ops)
 - [API文档](doc/API/API_index.html)
-- [示例Demo](demo)
-- [模型转换工具](tools/converter/README_CN.md)
-- [测试工具](doc/Tools_CN.md)
-- [Op 说明文档](doc/OpList.md)
-- [贡献代码](doc/Contributing_CN.md)
-- [模型压缩](tools/quantization/README_CN.md)
+- [示例](https://www.yuque.com/mnn/cn/demo_zoo)
 
-## 性能评测
-- [性能测试结果](doc/Benchmark_CN.md)
+## 工具
+- [测试工具](https://www.yuque.com/mnn/cn/tool_test)
+- [性能测试](https://www.yuque.com/mnn/cn/tool_benchmark)
+- [模型压缩](https://www.yuque.com/mnn/cn/tool_quantize)
 
-## 如何扩展
-- [添加自定义Op](doc/AddOp_CN.md)
-- [添加自定义Backend](doc/AddBackend_CN.md)
+## 如何修改
+- [自定义后端](https://www.yuque.com/mnn/cn/customize_backend)
+- [自定义算子](https://www.yuque.com/mnn/cn/customize_op)
+- [贡献代码](https://www.yuque.com/mnn/cn/contribute)
 
 ##  交流与反馈
-- [常见问题](doc/FAQ.md)
+- [常见问题](https://www.yuque.com/mnn/en/faq)
 
 扫描二维码加入钉钉讨论群。