• Automation framework: CK
  
• Development repository: ck-ml
  
• Source: GitHub
  
• Tags: profile,model-profile,qaic,qaic-profile
  
• Support for host OS: any
  
• Support for target OS: any
  
• Software detection plugin: compiler.glow.profile
  
• How to get the stable version from this portal via the CK client:
  
  pip install cbench
  cb download package:model-qaic-calibrate --version=3.0.0 --all
  ck install package:model-qaic-calibrate
  ck help package
  
• How to get the development version:
  
  pip install ck
  ck pull repo:ck-ml
  ck install package:model-qaic-calibrate
  ck help package
  
• Useful CK automation actions (CLI with the Python CK API and JSON IO):
  
  ck show env
  ck find env:{ID from above}
  ck virtual env:{ID from above}
  
• CLI and Python API: module:package
  
• Variations:
  • bs.1
  • bs.8
  • first.5
  • mlperf.option1
  • mlperf.option2
  • resnet50
  • ssd_resnet34

Profile ONNX or TF models using Glow

Prerequisites

TODO

ImageNet validation dataset

If you have the ImageNet validation dataset e.g. in /datasets/dataset-imagenet-ilsvrc2012-val, you can register it with CK as follows:

$dollar; echo "full" | ck detect soft:dataset.imagenet.val --extra_tags=full,ilsvrc2012 \
--full_path=/datasets/dataset-imagenet-ilsvrc2012-val/ILSVRC2012_val_00000001.JPEG

Examples

MLPerf ResNet50 ONNX with the first 5 images of ImageNet 2012 (quick test)

$dollar; ck install package --tags=profile,resnet50.onnx,first.5

MLPerf ResNet50 ONNX with calibration option 1

$dollar; ck install package --tags=profile,resnet50.onnx,mlperf.option1

MLPerf ResNet50 TF with calibration option 2

$dollar; ck install package --tags=profile,resnet50.tf,mlperf.option2

• 3.0.0 (2021-04-11)

• .cm/desc.json (3 bytes)
• .cm/info.json (306 bytes)
• .cm/meta.json (4602 bytes)
• README.md (870 bytes)
• install.sh (3567 bytes)