Deploy step failing for SSD object detection model while getting ready to inference model locally.

aws-samples / amazon-sagemaker-aws-greengrass-custom-object-detection-model

End to end code samples for training an object detection model in Amazon SageMaker using the built-in SSD algorithm and running it on AWS IoT Greengrass.

MIT No Attribution

39 stars 30 forks source link

Deploy step failing for SSD object detection model while getting ready to inference model locally. #14

Open MercyPrasanna opened 3 years ago

MercyPrasanna commented 3 years ago

Hi @angelarw, @zhreshold

I was following the below notebook to inference sagemaker object detection ssd model locally. I get an error reporting that commons module does not exist when executing the deploy step to make the model ready for deployment.

https://github.com/aws-samples/amazon-sagemaker-aws-greengrass-custom-object-detection-model/blob/master/training/03_local_inference_post_training.ipynb

Error:

makoit commented 2 years ago

@MercyPrasanna

Changes in the code are needed in https://github.com/zhreshold/mxnet-ssd because it is supported for python2 but we use python3.

change in symbol_builder.py: from common import multi_layer_feature, multibox_layer to from .common import multi_layer_feature, multibox_layer
changed xrange to range in symbol_builder.py

I made this changes and it worked for me.