change readme.md

[platform/upstream/caffeonacl.git] / acl_openailab / README.md

diff --git a/acl_openailab/README.md b/acl_openailab/README.md
index f399f17..526b2f8 100644 (file)
--- a/acl_openailab/README.md
+++ b/acl_openailab/README.md
@@ -1,17 +1,64 @@
-<img src="openailab.png" width = "308" height = "88" alt="OPEN AI LAB" align=center />
-
-The **Arm Compute Library** is a collection of low-level software functions optimized for Arm Cortex CPU and Arm Mali GPU architectures, targeted at a variety of use-cases including: **image processing**, **computer vision and machine learning**.  [![License](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
-
-**OPEN** AI LAB ported the [Caffe](http://caffe.berkeleyvision.org/) to support Arm Compute Library on  Rockchip RK3399. The target OS is Ubuntu 16.04. [![License](https://img.shields.io/badge/license-BSD-blue.svg)](LICENSE)
-
-# 1. Release Notes
-Please refer to [Release Notes](release_notes.md) for details
-
-# 2. Installation
-Please refer to [Installation](installation.md) for details
-
-# 3. Performance Report
-Please refer to [Performance Report](performance_report.pdf) for details
-
-# 4. User Manual 
-Please refer to [User Manual](user_manual.pdf) for details
+# CaffeOnACL\r
+[![License](https://img.shields.io/badge/license-BSD-blue.svg)](LICENSE)\r
+\r
+CaffeOnACL is a project that is maintained by **OPEN** AI LAB, it uses Arm Compute Library (NEON+GPU) to speed up [Caffe](http://caffe.berkeleyvision.org/) and provide utilities to debug, profile and tune application performance. \r
+\r
+The release version is 0.3.0, is based on [Rockchip RK3399](http://www.rock-chips.com/plus/3399.html) Platform, target OS is Ubuntu 16.04. Can download the source code from [OAID/CaffeOnACL](https://github.com/OAID/CaffeOnACL)\r
+\r
+* The ARM Computer Vision and Machine Learning library is a set of functions optimised for both ARM CPUs and GPUs using SIMD technologies. See also [Arm Compute Library](https://github.com/ARM-software/ComputeLibrary).\r
+* Caffe is a fast open framework for deep learning. See also [Caffe](https://github.com/BVLC/caffe).\r
+\r
+### Documents\r
+* [Installation instructions](https://github.com/OAID/CaffeOnACL/blob/master/acl_openailab/installation.md)\r
+* [User Manuals PDF](https://github.com/OAID/CaffeOnACL/blob/master/acl_openailab/user_manual.pdf)\r
+* [Performance Report PDF](https://github.com/OAID/CaffeOnACL/blob/master/acl_openailab/performance_report.pdf)\r
+\r
+### Arm Compute Library Compatibility Issues :\r
+There are some compatibility issues between ACL and Caffe Layers, we bypass it to Caffe's original layer class as the workaround solution for the below issues\r
+\r
+* Normalization in-channel issue\r
+* Tanh issue\r
+* Softmax supporting multi-dimension issue\r
+* Group issue\r
+\r
+Performance need be fine turned in the future\r
+\r
+# Release History\r
+The Caffe based version is [793bd96351749cb8df16f1581baf3e7d8036ac37](https://github.com/BVLC/caffe/tree/793bd96351749cb8df16f1581baf3e7d8036ac37).\r
+\r
+### Version 0.3.0 - Aug 26, 2017\r
+Support Arm Compute Library version 17.06 with 4 new layers added\r
+\r
+* Batch Normalization Layer\r
+* Direct convolution Layer\r
+* Locally Connect Layer\r
+* Concatenate layer\r
+\r
+\r
+### Version 0.2.0 - Jul 2, 2017\r
+\r
+Fix the issues:\r
+\r
+* Compatible with Arm Compute Library version 17.06\r
+* When OpenCL initialization fails, even if Caffe uses CPU-mode,it doesn't work properly.\r
+\r
+### Version 0.1.0 - Jun 2, 2017 \r
+   \r
+  Initial version supports 10 Layers accelerated by Arm Compute Library version 17.05 : \r
+\r
+* Convolution Layer\r
+* Pooling Layer\r
+* LRN Layer\r
+* ReLU Layer\r
+* Sigmoid Layer\r
+* Softmax Layer\r
+* TanH Layer\r
+* AbsVal Layer\r
+* BNLL Layer\r
+* InnerProduct Layer\r
+\r
+# 3 Issue Report\r
+Encounter any issue, please report on [issue report](https://github.com/OAID/CaffeOnACL/issues). Issue report should contain the following information :\r
+\r
+*  The exact description of the steps that are needed to reproduce the issue \r
+* The exact description of what happens and what you think is wrong \r