From 52a04f77bce966985246b1d72f5b5dac9a8f5276 Mon Sep 17 00:00:00 2001 From: liuqi Date: Sat, 24 Feb 2018 16:52:37 +0800 Subject: [PATCH] Change the README to english. --- README.md | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-) diff --git a/README.md b/README.md index 4ed0b2c4..27a05bda 100644 --- a/README.md +++ b/README.md @@ -1,30 +1,30 @@ # **MACE** - *Mobile(Mi) Accelerated Compute Engine Library* --- -小米自主研发的移动端神经网络加速引擎。 -*可加入用户邮件组 mace-users@xiaomi.com* - -## 简介 +## Introduction --- -**利用端侧的异构计算设备加速神经网络模型。** +**Accelerating Neural Network with Heterogeneous Computing Devices in the phone.** -目前支持的端侧计算设备包括：**CPU(NEON)/GPU/DSP**. +Supported Devices: **CPU(NEON)/GPU/DSP**. -## 架构 +## Architecture --- -采用Op组成的有向无环图的计算模式，使用**Tensor**对象存储所有数据并进行统一管理。 +- Use computational pattern of **DAG consisting of Ops**. +- **Tensor** objects manage all data. +- **Workspace** manage all **Tensors**. ## GPU --- -基于OpenCL 2.0实现，使用Image的存储格式优化内存访问和并行计算。 -针对不同Op的算法，设计对应的Image存储格式来优化内存访问。 +Use **Image** object to optimize memory access and parallel computing based on OpenCL 2.0. -下面是针对不同**Tensor**类型对应的Buffer和Image的格式。 -| Tensor类型 | Buffer格式 | Image格式 | 说明 | +Design the corresponding **Image** format to optimize memory access for different Op algorithm. +Each pixel of **Image** object contains four elements(e.g. RGBA). +The Following is **Buffer** and **Image** format for all **Tensors**. +| Tensor| Buffer| Image| Explanation| | --------- | :---------:|:--------:|:----:| -|Channel-Major Input/Output | NHWC | [W * (C+3)/4, N * H] | 默认输入输出的格式| -|Height-Major Input/Output | NHWC | [W * C, N * (H+3)/4] | Winograd Convolution所用格式| -|Width-Major Input/Output | NHWC | [(W+3)/4 * C, N * H] | Winograd Convolution所用格式| -|Convolution Filter | HWOI | [H * W * RoundUp<4>(I), (O+3)/4]|卷积核格式，尝试过[H*w*I, (O+3)/4]，两者性能没有区别| -|Depthwise Convlution Filter | HWIM | [H * W * M, (I+3)/4]|Depthwise卷积核格式| -|1-D Argument | W | [(W+3)/4, 1] | 一维参数格式，如Bias| \ No newline at end of file +|Channel-Major Input/Output | NHWC | [W * (C+3)/4, N * H] | Default Input/Output format| +|Height-Major Input/Output | NHWC | [W * C, N * (H+3)/4] | Winograd Convolution format| +|Width-Major Input/Output | NHWC | [(W+3)/4 * C, N * H] | Winograd Convolution format| +|Convolution Filter | HWOI | [H * W * RoundUp<4>(I), (O+3)/4]|Convolution filter format，There is no difference compared to [H*w*I, (O+3)/4]| +|Depthwise Convlution Filter | HWIM | [H * W * M, (I+3)/4]|Depthwise-Convolution filter format| +|1-D Argument | W | [(W+3)/4, 1] | 1D argument format, e.g. Bias| \ No newline at end of file -- GitLab