ctuning/object-detection-tflite.debian-9 repository overview

This container may be outdated. Please check CK-MLPerf docs⁠.

This repository contains Docker-ized instructions for the Collective Knowledge-powered object classification TFLite workloads in the MLPerf Inference benchmark: SSD-MobileNet (non-quantized, quantized), SSD-ResNet (TODO).

After pulling the latest image:

$ docker pull ctuning/object-detection-tflite.debian-9

you can use it e.g. as follows:

• run object detection using non-quantized SSD-MobileNet on 50 images:

$ docker run --rm ctuning/object-detection-tflite.debian-9 \
"ck run program:object-detection-tflite --dep_add_tags.weights=ssd-mobilenet,non-quantized --env.CK_BATCH_COUNT=50"

• step into bash:

$ docker run -it --rm ctuning/object-detection-tflite.debian-9 bash

For further information, see:

• build and run instructions⁠;
• benchmarking instructions⁠.