meta-digi/meta-digi-dey/dynamic-layers/stm32mpu-ai/recipes-samples/object-detection/models/onnx-models-coco-ssd-mobile...

28 lines
1.2 KiB
BlitzBasic

# Copyright (C) 2019, STMicroelectronics - All Rights Reserved
SUMMARY = "Create package containing COCO SSD MobileNetV1 models"
LICENSE = "Apache-2.0"
LIC_FILES_CHKSUM = "file://${COMMON_LICENSE_DIR}/Apache-2.0;md5=89aea4e17d99a7cacdbeed46a0096b10"
# This model is not available in the onnx/models repository and integrating the tf2onnx
# converter in the yocto workflow (to convert it from tflite) would require integrating
# tensorflow (not lite) and bazel as well, which is a pretty big undertaking.
# So, for now, we will simply include the converted model manually.
SRC_URI = " file://coco_ssd_mobilenet.onnx \
file://labels_coco_ssd_mobilenet_onnx.txt"
S = "${WORKDIR}"
do_configure[noexec] = "1"
do_compile[noexec] = "1"
do_install() {
install -d ${D}${prefix}/local/demo-ai/object-detection/models/coco_ssd_mobilenet
install -d ${D}${prefix}/local/demo-ai/object-detection/models/coco_ssd_mobilenet/testdata
# install coco ssd mobilenet model
install -m 0644 ${S}/label*.txt ${D}${prefix}/local/demo-ai/object-detection/models/coco_ssd_mobilenet/labels_coco_ssd_mobilenet_onnx.txt
install -m 0644 ${S}/*.onnx ${D}${prefix}/local/demo-ai/object-detection/models/coco_ssd_mobilenet/
}
FILES:${PN} += "${prefix}/local/"