• Automation framework: CK
  
• Development repository: ck-request-asplos18-mobilenets-armcl-opencl
  
• Source: GitHub
  
• Available command lines:
  
  • ck run program:mobilenets-armcl-opencl --cmd_key=default (META)
    
• Support for host OS: any
  
• Support for target OS: android, linux
  
• Tags: armcl,benchmark,mobilenet,mobilenets,float,f32,vopencl
  
• Template: image classification via ArmCL
  
• How to get the stable version via the client:
  
  pip install cbench
  cb download program:mobilenets-armcl-opencl --version=1.0.0 --all
  ck run program:mobilenets-armcl-opencl
• How to get the development version:
  
  pip install ck
  ck pull repo --url=https://github.com/dividiti/ck-request-asplos18-mobilenets-armcl-opencl
  ck run program:mobilenets-armcl-opencl
  

MobileNets-v1 program using Arm Compute Library

ImageNet classification and benchmarking using ArmCL and MobileNets-v1.

Requirements

ArmCL library

To build this program, you need ArmCL compiled with Graph API:

$ ck install package:lib-armcl-opencl-18.08 --env.USE_GRAPH=ON --env.USE_NEON=ON --extra_version=-graph

To build this program for Android you need to embed OpenCL kernels and select the target API as follows:

$ ck install package:lib-armcl-opencl-18.08 --env.USE_GRAPH=ON --env.USE_NEON=ON --extra_version=-graph \
--env.USE_EMBEDDED_KERNELS=ON --target_os=android23-arm64 [--env.DEBUG=ON]

NB: We have to embed kernels when building for Android as OpenCL kernel files are not copied to a remote device.

NB: Use --target_os=android23-arm64 to build for Android API 23 (v6.0 "Marshmallow") or similar.

TODO: For some reason only a debug build of the library can be used with this program on some Android devices. (When a release version is used, the program appears to get stuck at the graph preparation stage.)

Install pretrained and converted weights

Install one or more of the compatible MobileNet weights packages:

$ ck install package --tags=mobilenet,weights,npy

Install ImageNet validation dataset (50,000 images)

$ ck install package:imagenet-2012-aux
$ ck install package:imagenet-2012-val

Build

$ ck compile program:mobilenet-armcl-opencl [--target_os=android23-arm64] 

Run

$ ck run program:mobilenet-armcl-opencl [--target_os=android23-arm64] 

Benchmark

$ ck benchmark program:mobilenet-armcl-opencl [--target_os=android23-arm64] [--repetitions=10] [--dvdt_prof]

Program parameters

ArmCL parameters

Define a parameter by passing --env.<NAME>=<VALUE> to ck run program:mobilenets-armcl-opencl. See a common header for more details.

CK_CONVOLUTION_METHOD

• DEFAULT: use library defaults (possibly a mix of methods).
• GEMM: use GEMM-based convolutions.
• DIRECT: use direct convolutions.
• WINOGRAD: use Winograd convolutions (supported from v18.08; not applicable to MobileNets).

CK_LWS_TUNER_TYPE

• NONE: do not use any tuner (default).
• DEFAULT: use the default CLTuner (preferred).
• BIFROST: use the static BifrostTuner (may be deprecated).

CK_DATA_LAYOUT

• NCHW (default).
• NHWC (supported from v18.08).

• 1.0.0 (2019-10-08)

• .cm/desc.json (3 bytes)
• .cm/info.json (319 bytes)
• .cm/meta.json (3714 bytes)
• README.md (2594 bytes)
• benchmark.cpp (3430 bytes)
• mobilenet.cpp (12198 bytes)
• preprocess-next.py (2002 bytes)