Centernet_on_mobile does not work with default Notebook

Prerequisites

Please answer the following questions for yourself before submitting an issue.

• I am using the latest TensorFlow Model Garden release and TensorFlow 2.
• I am reporting the issue to the correct repository. (Model Garden official or research directory)
• I checked to make sure that this issue has not already been filed.

1. The entire URL of the file you are using

https://github.com/tensorflow/models/blob/master/research/object_detection/colab_tutorials/centernet_on_device.ipynb

2. Describe the bug

When trying to run the cell:
`%%bash

Export the intermediate SavedModel that outputs 10 detections & takes in an

image of dim 320x320.

Modify these parameters according to your needs.

python models/research/object_detection/export_tflite_graph_tf2.py
--pipeline_config_path=centernet_mobilenetv2_fpn_od/pipeline.config
--trained_checkpoint_dir=centernet_mobilenetv2_fpn_od/checkpoint
--output_directory=centernet_mobilenetv2_fpn_od/tflite
--centernet_include_keypoints=false
--max_detections=10
--config_override="
model{
center_net {
image_resizer {
fixed_shape_resizer {
height: 320
width: 320
}
}
}
}" I get the following bug: Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/mobilenet_v2/mobilenet_v2_weights_tf_dim_ordering_tf_kernels_1.0_224_no_top.h5
9406464/9406464 [==============================] - 0s 0us/step
2022-06-27 12:55:49.027224: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
WARNING:tensorflow:input_shape is undefined or non-square, or rows is not in [96, 128, 160, 192, 224]. Weights for input shape (224, 224) will be loaded as the default.
W0627 12:55:49.031333 140636195452800 mobilenet_v2.py:303] input_shape is undefined or non-square, or rows is not in [96, 128, 160, 192, 224]. Weights for input shape (224, 224) will be loaded as the default.
Traceback (most recent call last):
File "models/research/object_detection/export_tflite_graph_tf2.py", line 160, in
app.run(main)
File "/usr/local/lib/python3.7/dist-packages/absl/app.py", line 312, in run
_run_main(main, args)
File "/usr/local/lib/python3.7/dist-packages/absl/app.py", line 258, in _run_main
sys.exit(main(argv))
File "models/research/object_detection/export_tflite_graph_tf2.py", line 156, in main
FLAGS.centernet_include_keypoints, FLAGS.keypoint_label_map_path)
File "/usr/local/lib/python3.7/dist-packages/object_detection/export_tflite_graph_lib_tf2.py", line 364, in export_tflite_model
status.assert_existing_objects_matched()
File "/usr/local/lib/python3.7/dist-packages/tensorflow/python/training/tracking/util.py", line 850, in assert_existing_objects_matched
f"Found {num_unused_python_objects} Python objects that were "
AssertionError: Found 3 Python objects that were not bound to checkpointed values, likely due to changes in the Python program. Showing 3 of 3 unmatched objects: [<tf.Variable 'conv2d_2/kernel:0' shape=(3, 3, 64, 32) dtype=float32, numpy=
array([[[[ 7.32026994e-04, -6.46058545e-02, 5.72565570e-02, ...,
1.02764368e-03, -3.61360721e-02, -6.56415075e-02],
[-5.33955097e-02, -3.59095559e-02, 8.10381547e-02, ...,
5.55812344e-02, -3.71528491e-02, -2.78462768e-02],
[-1.86755434e-02, -5.61659560e-02, 7.76920691e-02, ...,
1.56775936e-02, -4.45304327e-02, 4.94885221e-02],
...,
[-3.47078666e-02, 1.01511478e-02, -2.71509886e-02, ...,
-7.88835138e-02, 6.08911589e-02, -4.40879464e-02],
[ 4.69631627e-02, 7.06758350e-03, -2.16422901e-02, ...,
6.20872453e-02, -6.10724092e-05, 1.42914876e-02],
[ 4.31788042e-02, -2.55548581e-02, 2.60739326e-02, ...,
5.49626723e-02, 2.70691141e-02, 8.55640322e-03]],

    [[ 5.65723330e-03, -5.94574213e-02, -5.61843738e-02, ...,
      -1.01557970e-02, -7.79245347e-02, -8.10062289e-02],
     [ 5.65348342e-02,  4.03069481e-02, -2.13610344e-02, ...,
       5.70722446e-02,  7.32235983e-02,  3.48072872e-02],
     [ 7.60557428e-02, -3.27609181e-02,  1.05622783e-02, ...,
      -1.04101077e-02, -2.46028304e-02,  2.74680033e-02],
     ...,
     [ 4.61962894e-02, -3.43475938e-02, -3.56415138e-02, ...,
       1.81821212e-02, -7.35602602e-02, -6.83561787e-02],
     [ 4.92095947e-03,  6.42872825e-02, -4.98419814e-02, ...,
      -5.17259464e-02,  9.17863846e-03,  7.47016817e-03],
     [-7.82151669e-02, -1.36788115e-02,  6.27572909e-02, ...,
      -3.60929966e-03, -4.43668365e-02,  3.90753746e-02]],

    [[-2.11395323e-04, -7.61388764e-02, -4.05194983e-02, ...,
      -3.71707901e-02, -5.36263585e-02,  1.99466944e-02],
     [-3.44925746e-02,  7.80378804e-02,  3.90290022e-02, ...,
       7.16263875e-02,  6.25740066e-02, -3.08517367e-03],
     [-8.32480416e-02, -2.23641023e-02,  3.21296826e-02, ...,
       5.88224307e-02, -3.32544446e-02,  2.89830342e-02],
     ...,
     [-4.97186184e-02,  6.37233034e-02,  8.02874938e-02, ...,
      -8.19565877e-02, -6.61344975e-02,  2.07980722e-03],
     [-1.57658830e-02, -4.38169241e-02, -2.27425098e-02, ...,
       9.75948572e-03,  8.00549760e-02, -5.84334135e-03],
     [-4.08791713e-02,  7.14382306e-02,  7.62904361e-02, ...,
       7.26937130e-02, -7.98386335e-03,  7.08226189e-02]]],


   [[[ 1.56138316e-02,  3.96466032e-02, -7.40392208e-02, ...,
      -4.74337488e-03, -4.30962257e-02, -2.61694193e-02],
     [ 3.79069075e-02, -4.06743698e-02,  8.97413492e-03, ...,
       6.42007962e-02,  5.59016541e-02,  7.84287080e-02],
     [-7.64534920e-02,  5.37436679e-02, -4.57442626e-02, ...,
      -1.34410709e-03,  1.68981403e-03,  6.04746118e-02],
     ...,
     [-1.13270059e-02, -8.16817507e-02,  8.20283368e-02, ...,
      -1.37368217e-02, -6.83582425e-02, -7.36899376e-02],
     [ 4.21300754e-02, -7.00433701e-02, -4.81314883e-02, ...,
       8.11765566e-02,  2.57344842e-02, -7.75993466e-02],
     [-1.75237432e-02, -6.91977739e-02,  1.21745095e-02, ...,
      -6.27882928e-02, -5.56558967e-02,  5.89358434e-02]],

    [[ 4.77465466e-02,  2.62621045e-02, -1.48655549e-02, ...,
       3.99657711e-02, -7.61527643e-02,  1.30702481e-02],
     [ 7.39316121e-02, -3.39320302e-02,  6.43393025e-02, ...,
       4.58922237e-03,  3.81932259e-02, -1.96603537e-02],
     [ 6.50526956e-02, -9.84913111e-03,  8.10517892e-02, ...,
      -7.29359984e-02,  1.09341741e-02, -1.16719604e-02],
     ...,
     [ 6.76664189e-02, -6.34652376e-03, -6.66577443e-02, ...,
       8.32766667e-02,  7.84228072e-02, -4.49749082e-03],
     [ 5.42106703e-02, -5.19617647e-03, -1.96215138e-02, ...,
       8.20146278e-02,  4.19336557e-03,  2.69444957e-02],
     [-5.74513525e-03,  3.80031690e-02,  2.24485248e-03, ...,
      -1.23658404e-02, -2.04354301e-02,  2.50922218e-02]],

    [[-7.83682466e-02,  4.37875167e-02,  9.48166847e-03, ...,
       6.63023219e-02, -5.52075319e-02,  4.66249362e-02],
     [-3.51882987e-02, -6.43727034e-02, -1.46498457e-02, ...,
      -5.31300716e-02,  4.55096588e-02, -1.87502056e-03],
     [ 8.17356482e-02,  7.50396401e-03, -1.97547674e-02, ...,
      -5.20819426e-03, -7.90962204e-02,  7.45240226e-02],
     ...,
     [-1.97505131e-02,  1.78856030e-02, -3.85709032e-02, ...,
       2.65328661e-02, -6.93191513e-02, -3.61771584e-02],
     [ 7.73564801e-02, -5.54815531e-02, -9.33017582e-03, ...,
       2.37973332e-02,  5.86604103e-02,  8.88586044e-03],
     [-4.86270189e-02, -6.42125010e-02,  6.05420247e-02, ...,
       5.71920723e-03,  2.86288857e-02, -6.11796230e-03]]],


   [[[ 7.41044059e-02, -4.12326679e-02, -2.70548463e-02, ...,
      -6.36980757e-02, -5.91927990e-02,  1.30941495e-02],
     [ 6.07319102e-02, -4.51031327e-02, -2.36622691e-02, ...,
       3.63916755e-02,  5.55339828e-02, -4.76327352e-02],
     [-5.37601858e-03,  2.65799463e-04, -7.43721575e-02, ...,
      -7.10965246e-02,  1.83032528e-02, -4.16711979e-02],
     ...,
     [-2.80845985e-02, -3.20065431e-02,  8.08035210e-02, ...,
       7.58523419e-02, -6.93654642e-02, -3.25286612e-02],
     [-7.48524293e-02,  1.50763988e-03, -1.09771490e-02, ...,
       1.86863914e-02, -2.84503512e-02,  1.95456743e-02],
     [ 1.70392990e-02, -2.12874264e-03,  2.54323855e-02, ...,
      -3.93110327e-02,  3.48169431e-02,  1.92938447e-02]],

    [[-7.45099634e-02,  5.15094176e-02,  6.66167811e-02, ...,
      -6.24125414e-02,  8.07455257e-02, -1.93100199e-02],
     [-5.93174696e-02, -4.16195020e-02, -2.36678123e-03, ...,
       5.29523119e-02, -5.49765825e-02, -4.53216434e-02],
     [ 2.71362439e-02, -3.33693624e-02, -1.44081116e-02, ...,
      -3.71100903e-02, -5.05258851e-02,  4.56229523e-02],
     ...,
     [-1.36879832e-03,  3.82750258e-02,  1.59386024e-02, ...,
      -6.80658221e-02, -6.28489479e-02, -2.96099782e-02],
     [ 8.02307948e-02, -7.70650357e-02, -1.23209134e-02, ...,
       2.48978958e-02,  4.34805229e-02, -7.78292045e-02],
     [-5.63441329e-02,  6.47042766e-02,  8.80495459e-03, ...,
       4.20342311e-02, -1.69240832e-02,  1.53372437e-03]],

    [[-1.39151439e-02,  4.11764532e-03, -3.86212282e-02, ...,
       1.15843192e-02, -3.03387642e-04, -2.15180330e-02],
     [ 6.42243847e-02,  6.81555495e-02, -4.60349545e-02, ...,
      -6.42998442e-02,  4.67669740e-02,  7.98614249e-02],
     [ 4.21014056e-02,  8.31046328e-02,  1.52886510e-02, ...,
      -7.85305351e-02,  1.43661276e-02,  3.15602869e-03],
     ...,
     [-8.27382058e-02, -1.62731037e-02,  6.06246367e-02, ...,
       4.28032875e-03,  1.66793689e-02, -6.71548247e-02],
     [ 7.46393427e-02,  5.06779626e-02, -6.37319684e-02, ...,
      -7.99990520e-02,  4.97202948e-02, -3.42036299e-02],
     [ 3.89334559e-02, -5.70040755e-02,  2.33808532e-02, ...,
       2.15160474e-02,  6.16992489e-02, -6.89772516e-03]]]],
  dtype=float32)>, <tf.Variable 'conv2d_4/kernel:0' shape=(3, 3, 32, 24) dtype=float32, numpy=

array([[[[-6.96186125e-02, 3.08893546e-02, 4.33606580e-02, ...,
-8.37306753e-02, 9.01646540e-02, 5.60158417e-02],
[-1.64597705e-02, -5.95558994e-02, 6.17236719e-02, ...,
7.82588944e-02, 5.45480773e-02, -1.06570385e-01],
[-1.05534911e-01, 2.43080482e-02, 9.10778865e-02, ...,
-1.44298226e-02, -8.49713385e-03, 5.72649315e-02],
...,
[-3.08786333e-03, 1.53949782e-02, -1.05845928e-02, ...,
9.92716923e-02, 1.03756852e-01, -8.15490484e-02],
[-9.10099074e-02, -1.91824511e-02, 6.04171827e-02, ...,
9.06801298e-02, -1.04863375e-01, 2.78138742e-02],
[-2.23484933e-02, -3.93129736e-02, -3.07530165e-04, ...,
-1.07519127e-01, -7.72145540e-02, -7.04452246e-02]],

    [[ 7.29673728e-02, -4.22494411e-02, -4.86789122e-02, ...,
       6.43370226e-02, -6.42939135e-02,  2.27802917e-02],
     [ 1.84721276e-02,  4.72623110e-03,  3.17010805e-02, ...,
       4.25498709e-02,  7.35374317e-02,  8.24982673e-03],
     [ 2.79041454e-02,  9.34241712e-03, -5.13544679e-03, ...,
      -4.58292514e-02, -9.32499394e-02, -8.31490904e-02],
     ...,
     [-1.35802925e-02,  8.55067596e-02, -8.38937834e-02, ...,
       3.72737721e-02,  1.58094540e-02, -7.60663450e-02],
     [-7.24469647e-02,  5.81275448e-02, -5.45617305e-02, ...,
      -3.06949764e-03,  7.27384537e-03,  3.83527055e-02],
     [ 1.03801258e-01,  9.72040370e-02,  8.61172155e-02, ...,
       4.89479378e-02, -7.13758543e-02,  7.40995035e-02]],

    [[ 3.16187739e-03, -2.93501019e-02,  1.07661031e-01, ...,
       1.04346432e-01, -7.20791593e-02,  8.25662240e-02],
     [ 1.08621992e-01,  5.75260594e-02, -8.34193826e-03, ...,
      -1.02050155e-01,  8.74772295e-02, -4.45634276e-02],
     [-1.04781479e-01,  5.00401780e-02, -7.48666227e-02, ...,
       9.94708315e-02,  3.31482962e-02,  1.30900666e-02],
     ...,
     [-7.11073726e-02,  7.48594627e-02,  2.12256089e-02, ...,
       4.24627960e-03, -1.23680085e-02, -8.20651650e-04],
     [ 3.89846787e-02, -1.80026069e-02,  1.71592012e-02, ...,
      -1.04135014e-01,  5.11140451e-02,  1.96356699e-02],
     [ 4.44334969e-02,  1.04438938e-01,  8.81219879e-02, ...,
       6.73528835e-02,  8.49354640e-02, -5.21814339e-02]]],


   [[[ 4.78038117e-02, -9.05216560e-02,  9.60739702e-03, ...,
       8.92607793e-02, -6.63148835e-02,  5.91619313e-03],
     [-1.03514180e-01,  6.47801086e-02,  1.14756897e-02, ...,
       3.51345167e-02, -3.12762558e-02,  4.04082015e-02],
     [ 9.60062668e-02,  8.62717628e-03, -7.06970096e-02, ...,
       6.76457584e-03,  1.00657344e-05,  8.08915570e-02],
     ...,
     [ 9.37923267e-02,  7.91135207e-02, -8.49928856e-02, ...,
      -1.40290335e-02,  3.77522334e-02, -3.61695439e-02],
     [-7.04932213e-03,  3.83533612e-02,  6.25432953e-02, ...,
       2.25975141e-02,  7.06030950e-02,  9.01883841e-03],
     [-7.03543276e-02, -6.15260415e-02,  7.90153965e-02, ...,
      -9.09473971e-02, -8.73929262e-03, -2.58195400e-02]],

    [[ 5.74920401e-02, -1.01080008e-01, -4.85088341e-02, ...,
      -3.96420732e-02,  1.45243034e-02,  1.59256831e-02],
     [ 8.49615559e-02, -4.71835993e-02, -2.36850977e-02, ...,
       8.81597400e-04, -4.84608412e-02, -2.53227577e-02],
     [-5.10167032e-02,  8.96194577e-03,  4.18808535e-02, ...,
      -5.67450263e-02,  1.07488610e-01,  1.01591177e-01],
     ...,
     [-5.77555262e-02, -2.34076604e-02,  5.65244630e-02, ...,
       5.55771217e-02, -6.38792068e-02,  3.91295478e-02],
     [-5.39851412e-02, -8.11582953e-02,  9.85756516e-03, ...,
       3.71588767e-03,  5.28684780e-02, -4.66035455e-02],
     [-7.83324242e-03, -7.56454170e-02,  3.57057229e-02, ...,
      -3.80651653e-03,  5.99108636e-03, -1.69722363e-02]],

    [[ 1.05387755e-01,  9.37174335e-02, -8.04332495e-02, ...,
       1.08015761e-02, -7.83426389e-02, -6.49992973e-02],
     [-5.63945211e-02, -6.12401254e-02, -1.06169641e-01, ...,
       3.08536366e-02, -3.97268012e-02,  6.46192357e-02],
     [-9.85943824e-02, -4.35892195e-02, -2.14079171e-02, ...,
       6.58106580e-02, -8.68956745e-02, -6.17109984e-03],
     ...,
     [ 9.39257666e-02,  9.93669853e-02,  5.10205850e-02, ...,
      -4.63575125e-05, -8.30020905e-02, -1.05410151e-01],
     [ 1.00582615e-02,  2.59639099e-02,  6.64621964e-02, ...,
      -1.17293000e-03, -1.02259725e-01,  6.08614087e-03],
     [ 7.88109824e-02, -8.12843591e-02, -5.72104380e-02, ...,
       3.09819356e-02, -1.81559026e-02, -1.07827708e-02]]],


   [[[ 6.98380694e-02, -8.78748298e-02, -9.24986154e-02, ...,
      -2.54325047e-02,  6.23675212e-02,  1.00935392e-01],
     [ 1.05525188e-01,  3.27825919e-02, -1.25111341e-02, ...,
      -3.69934216e-02,  7.23836571e-03,  1.80671290e-02],
     [ 7.97397718e-02,  5.67120686e-02,  1.41721815e-02, ...,
      -6.91904873e-03,  6.21857420e-02, -3.95446792e-02],
     ...,
     [ 3.06623653e-02,  1.00788780e-01,  6.39619604e-02, ...,
      -2.45221630e-02, -2.19101086e-02, -3.22953612e-03],
     [-1.03909716e-01, -1.32028386e-02, -4.10056040e-02, ...,
      -6.23378977e-02,  9.52673480e-02, -5.07372394e-02],
     [ 2.13041827e-02,  1.83184370e-02, -7.72977769e-02, ...,
      -1.84653103e-02, -1.09036185e-01,  3.95091251e-02]],

    [[-1.05999410e-02,  9.40769091e-02, -7.83783272e-02, ...,
      -6.93801045e-03,  9.23362151e-02,  9.30364951e-02],
     [-7.93510228e-02, -3.43228132e-02,  1.07420586e-01, ...,
       2.43902281e-02,  1.16922557e-02, -2.70873383e-02],
     [ 6.71553090e-02, -1.33232549e-02,  2.85041705e-02, ...,
      -9.83355939e-02,  7.50312135e-02,  8.64572600e-02],
     ...,
     [-5.07462770e-03, -1.08758651e-01,  6.81450441e-02, ...,
       1.66752860e-02, -7.19465464e-02,  5.22499755e-02],
     [-6.93949163e-02,  1.00683577e-01,  2.10750774e-02, ...,
      -4.77043130e-02,  1.01492457e-01, -1.49386972e-02],
     [-3.04999575e-02, -4.35249507e-03,  5.25384769e-02, ...,
       7.93131441e-03, -1.04433469e-01,  8.71478245e-02]],

    [[-2.96992064e-03, -9.57195610e-02, -8.07940513e-02, ...,
       3.78137901e-02, -1.08848810e-01, -1.07406512e-01],
     [-2.43262574e-02, -9.21401232e-02,  2.42572352e-02, ...,
       6.45631775e-02,  3.70400622e-02,  5.83791807e-02],
     [-8.55684355e-02, -1.15403906e-02,  1.03023879e-01, ...,
      -1.64935067e-02, -8.46476853e-02, -1.90330818e-02],
     ...,
     [ 4.72837016e-02, -2.22670436e-02, -8.06952044e-02, ...,
       2.63160542e-02, -1.51020661e-02, -2.87870690e-02],
     [-1.54100135e-02, -7.00154454e-02, -2.74903178e-02, ...,
       1.82549432e-02,  9.77394655e-02,  9.21355188e-03],
     [ 6.74366429e-02, -7.71830529e-02,  7.50297830e-02, ...,
      -9.76691023e-02,  1.51817203e-02, -5.66378236e-03]]]],
  dtype=float32)>, <tf.Variable 'conv2d_6/kernel:0' shape=(3, 3, 24, 24) dtype=float32, numpy=

array([[[[ 0.08599023, -0.04533422, -0.04254126, ..., 0.11302692,
-0.03133832, 0.01305629],
[-0.0556149 , -0.10821456, 0.04288476, ..., -0.016245 ,
0.09893914, -0.04627853],
[ 0.021794 , 0.06245694, -0.00917228, ..., -0.08639474,
0.07575055, 0.06011017],
...,
[ 0.06592857, 0.07378992, -0.07895294, ..., 0.0945333 ,
-0.09966353, 0.01914709],
[-0.07417758, -0.10152224, -0.08046505, ..., 0.07279224,
0.06833985, 0.11264979],
[ 0.02052207, -0.01488113, 0.08209715, ..., 0.06637131,
0.10581297, -0.01998039]],

    [[-0.04365531, -0.09509368,  0.11498032, ..., -0.10188408,
       0.01250995,  0.00809995],
     [-0.03705225,  0.00895362,  0.07704405, ..., -0.10366832,
      -0.04790037,  0.01073112],
     [ 0.04538529, -0.07509759,  0.05771219, ..., -0.07512055,
      -0.09624924, -0.03582856],
     ...,
     [ 0.06350049,  0.07504167, -0.05847465, ...,  0.03909988,
       0.11501703, -0.01109319],
     [-0.09677709,  0.04563666, -0.02321906, ...,  0.07438303,
      -0.07104559,  0.00217865],
     [ 0.1162475 ,  0.08059127, -0.06234303, ...,  0.01210871,
      -0.02910072, -0.07536888]],

    [[-0.11539479,  0.1060719 ,  0.05812388, ..., -0.03612803,
      -0.06755395, -0.08651754],
     [-0.04313466, -0.029396  , -0.01694685, ...,  0.02235831,
       0.10438592,  0.07718193],
     [ 0.00097219,  0.05851323,  0.03490411, ..., -0.00636058,
       0.11606646,  0.06524048],
     ...,
     [-0.00353374, -0.07650431,  0.06702108, ...,  0.01945465,
       0.06410234,  0.08320887],
     [-0.05022384, -0.06732228, -0.02821372, ..., -0.07599302,
       0.01095106, -0.0045016 ],
     [-0.06140565,  0.01708234, -0.08480848, ..., -0.05199617,
      -0.10550337, -0.09903585]]],


   [[[-0.06969035,  0.06000992, -0.10664372, ...,  0.07174978,
       0.1000635 , -0.04401154],
     [-0.01003904, -0.07922419,  0.02758362, ...,  0.07465743,
       0.08611778, -0.08105084],
     [ 0.02601161, -0.02289852, -0.0379418 , ...,  0.09313001,
      -0.04988813,  0.05673233],
     ...,
     [ 0.01455951, -0.07434285,  0.01565044, ...,  0.10239203,
      -0.08436647, -0.04102403],
     [ 0.0104328 , -0.05404482,  0.08538812, ..., -0.04182867,
      -0.02013201, -0.02101748],
     [-0.02478645, -0.0764901 , -0.07568152, ..., -0.03867862,
      -0.01241989,  0.08824646]],

    [[ 0.07987603, -0.05483631, -0.11702188, ...,  0.03601704,
       0.05851723,  0.06360755],
     [-0.11355839, -0.10092273,  0.02904665, ..., -0.02572539,
      -0.00137469, -0.057562  ],
     [ 0.08275711, -0.05495244, -0.02587297, ..., -0.06793822,
       0.08480457,  0.00379105],
     ...,
     [-0.05961942, -0.03399573, -0.07033213, ...,  0.03665494,
       0.01451214, -0.07146159],
     [-0.01094893, -0.11483898, -0.07840778, ...,  0.06974404,
      -0.02518418,  0.10362994],
     [-0.01470638,  0.10088163,  0.06474959, ...,  0.0030736 ,
      -0.07322711, -0.04664299]],

    [[ 0.11283394,  0.06285908,  0.07415528, ..., -0.106759  ,
      -0.0627305 ,  0.05050515],
     [-0.07335878, -0.0110747 , -0.02585484, ...,  0.08675175,
       0.05679605,  0.07723866],
     [ 0.10995717, -0.0446708 ,  0.07399734, ..., -0.07755931,
       0.09459584,  0.10548358],
     ...,
     [-0.03491943, -0.08102164, -0.02385981, ..., -0.06765938,
       0.02489876, -0.03668824],
     [ 0.01632432, -0.00325862, -0.01031623, ..., -0.11587231,
      -0.11626981, -0.03502592],
     [ 0.04202921, -0.03054334,  0.01374257, ..., -0.09626315,
      -0.02604263, -0.05408505]]],


   [[[-0.04666702,  0.08656212,  0.07679554, ..., -0.0661195 ,
      -0.0833699 , -0.01198196],
     [-0.01587186, -0.1129812 , -0.09766321, ...,  0.06264401,
       0.04261594, -0.06040349],
     [-0.06727182,  0.04031301, -0.11635529, ..., -0.08418209,
       0.08475371,  0.09542347],
     ...,
     [ 0.00607643, -0.00253856, -0.09863922, ..., -0.02235598,
      -0.07438292,  0.09799305],
     [-0.10496243,  0.03628122,  0.10159124, ..., -0.06678902,
       0.00784139, -0.06914005],
     [ 0.06828318,  0.0603999 ,  0.03736115, ...,  0.01658658,
      -0.10790116,  0.02782493]],

    [[ 0.09665345, -0.00914921,  0.04845747, ...,  0.09723713,
       0.01673449,  0.0398009 ],
     [-0.07808128,  0.04481471, -0.07592319, ...,  0.0111608 ,
      -0.11099119, -0.09936887],
     [-0.07213995,  0.1135022 , -0.02246975, ..., -0.05470144,
      -0.09145998,  0.09958873],
     ...,
     [ 0.03735975, -0.03798223, -0.00858913, ..., -0.07289623,
       0.09257367,  0.08717654],
     [-0.08839332,  0.01774638, -0.08159545, ..., -0.02916528,
       0.00076696, -0.04589005],
     [ 0.10006448, -0.08874221,  0.10163201, ...,  0.09024823,
       0.10095913,  0.04654611]],

    [[ 0.04319265, -0.08179846, -0.03158928, ...,  0.10363428,
      -0.06356049, -0.03064433],
     [-0.07858384,  0.09244021,  0.11619588, ...,  0.10436533,
      -0.00040054, -0.09745093],
     [-0.0560239 ,  0.04749753,  0.11004435, ...,  0.04017404,
       0.11347028,  0.07855082],
     ...,
     [-0.02334446,  0.06316862, -0.03868822, ...,  0.05266895,
      -0.09784517,  0.11078187],
     [-0.02890265,  0.058161  , -0.0027514 , ...,  0.09047166,
       0.04920878,  0.05365493],
     [ 0.06582298,  0.10608193,  0.06625228, ...,  0.07816046,
      -0.09903563,  0.07565663]]]], dtype=float32)>]`

3. Steps to reproduce

Start a new colab Notebook and run all the cells.
https://github.com/tensorflow/models/blob/master/research/object_detection/colab_tutorials/centernet_on_device.ipynb

4. Expected behavior

The Notebook runs as expected and outputs bboxes.

5. Additional context

---

6. System information

• OS Platform and Distribution (e.g., Linux Ubuntu 16.04):
• Mobile device name if the issue happens on a mobile device:
• TensorFlow installed from (source or binary): binary (pip install tf-nightly)
• TensorFlow version (use command below): 2.9.1
• Python version:
• Bazel version (if compiling from source):
• GCC/Compiler version (if compiling from source):
• CUDA/cuDNN version:
• GPU model and memory:

Start with research/object_detection/colab_tutorials/centernet_on_device.ipynb and reproduce the export cell using export_tflite_graph_tf2.py. Trace the failure into export_tflite_graph_lib_tf2.py and compare the model variables with the checkpoint; done means the notebook cell completes and produces the requested TFLite export without the unmatched-object assertion.