[Bug Report]Error in example notebook Amazon_Tensorflow_Object_Detection

**Link to the notebook**
[amazon-sagemaker-examples](https://github.com/aws/amazon-sagemaker-examples/tree/4ca4fbb9246739693110d83f00c952e105cdd704)/[ archived](https://github.com/aws/amazon-sagemaker-examples/tree/4ca4fbb9246739693110d83f00c952e105cdd704/%20archived)
/Amazon_Tensorflow_Object_Detection.ipynb
**Describe the bug**
TypeError: the JSON object must be str, bytes or bytearray, not dict
thrown using default parameters/selections

at
normalized_boxes, classes_names, confidences = parse_response(query_response)

**To reproduce**
A clear, step-by-step set of instructions to reproduce the bug.
open notebook
tensorflow-od1-ssd-resnet50-v1-fpn-1024x1024-coco17-tpu-8-SDK.ipynb
step through the notebook executing each line

**Logs**
If applicable, add logs to help explain your problem.
You may also attach an `.ipynb` file to this issue if it includes relevant logs or output.
[tensorflow-od1-ssd-resnet50-v1-fpn-1024x1024-coco17-tpu-8-SDK.ipynb.txt](https://github.com/user-attachments/files/17196462/tensorflow-od1-ssd-resnet50-v1-fpn-1024x1024-coco17-tpu-8-SDK.ipynb.txt)

Open archived/Amazon_Tensorflow_Object_Detection.ipynb and the linked tensorflow-od1-ssd-resnet50-v1-fpn-1024x1024-coco17-tpu-8-SDK notebook. Run the notebook with its default selections and inspect parse_response(query_response) at the reported failing line. Done means the default flow completes without the JSON TypeError.

Written by the indexing model from the issue text.