"ValueError: numpy.ndarray size changed " while trying Intel lpot in tensorflow model

Question

While trying out the Intel Low Precision Optimization Tool in tensorflow model, getting some value error.

Please find the command I tried below:

# The cmd of running ssd_resnet50_v1
bash run_tuning.sh --config=ssd_resnet50_v1.yaml --input_model=/tmp/ssd_resnet50_v1_fpn_shared_box_predictor_640x640_coco14_sync_2018_07_03/frozen_inference_graph.pb --output_model=./tensorflow-ssd_resnet50_v1-tune.pb

By running, I am getting the below error:

import pycocotools._mask as _mask File "pycocotools/_mask.pyx", line 1, in init pycocotools._mask ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject

Providing the git hub link I followed: https://github.com/intel/neural-compressor/tree/master/examples/tensorflow/object_detection

Answer 1

Please try to upgrade numpy version. Command is given below:

pip install --upgrade numpy

I had this same issue and what I did was upgrade numpy, which resolved the issue.

Thanks