Angelo Delli Santi

Hello bsrkhan. We build a custom numpy to run using openblas on our camera. As you see in the [dockerfile](https://github.com/AxisCommunications/acap-computer-vision-sdk-examples/blob/main/object-detector-python/Dockerfile#L15) we currently copy the built numpy manually in the example....

It seems --no-dependencies doesn't exist anymore, use `pip install --no-deps pandas` instead

Hi Unfortunately, we didn't find a work around yet. We saw that by adding `RUN touch /usr/lib/python3.8/site-packages/numpy-1.17.3.egg-info` to the Dockerfile it seems you can "trick" pip to recognize the presence...

We'll update this issue, when we'll be able to improve the pip packaging

@bsrkhan I just would like to mention that this issue have not been prioritized, so even if we plan to address this problem, unfortunately we can't give an ETA for...

Hey @bsrkhan Maybe you can take a look at [this issue](https://github.com/AxisCommunications/acap-computer-vision-sdk/issues/32#issuecomment-1375680692) @daniel-falk showed how you can install pandas. This will work however only on aarch64 devices.

HI @akash-syook I will transfer this issue to the example repository

Hi @HardcoreBudget Here you can find some guidelines on how to post-process yolo output: https://github.com/AxisCommunications/axis-model-zoo/discussions/45 Not sure if I understood the second part, what version was faster? And what weights...

It is most likely that your version of SSD mobilenetv2 was fine, but quantized by channel. That can have a large impact on latency. See the [documentation](https://developer.axis.com/computer-vision/computer-vision-on-device/quantization) for more info...

Hi, That version of ssd mobilenet comes from here: https://coral.ai/models/object-detection/ This should be its config file for TensorFlow: https://github.com/tensorflow/models/blob/master/research/object_detection/samples/configs/ssd_mobilenet_v2_quantized_300x300_coco.config