Fpenet custom dataset train failed

Please provide the following information when requesting support.

• Network Type fpenet
• TLT Version 3.22.05
• Training spec file
experiment_spec_pig.yaml (2.3 KB)
I want to use 27 custom keypoints train fpenet
deadpig_weight_1660549603537_1.pts (249 Bytes)
I modified the dataset to be the same as afw.
After that, I modified function convert_dataset and convert data successful

        if key_points == 80:
             for num, keypoint in enumerate(image_keypoints):
                 annotations["P{}x".format(num+1)] = keypoint[0]
                 annotations["P{}y".format(num+1)] = keypoint[1]

             # fill in dummy keypoints for keypoints 69 to 80
             for num in range(69, 81, 1):
                 annotations["P{}x".format(num)] = image_keypoints[0][0]
                 annotations["P{}y".format(num)] = image_keypoints[0][1]
                 annotations["P{}occluded".format(num)] = True
         elif key_points==10:
             key_id = 1
             for num, keypoint in enumerate(image_keypoints):
                 # change to 10-points dataset:
                 if (num+1) in [1, 9, 17, 20, 25, 39, 45, 34, 49, 55]:
                     annotations["P{}x".format(key_id)] = keypoint[0]
                     annotations["P{}y".format(key_id)] = keypoint[1]
                     key_id += 1
         elif key_points == 27:
             for num, keypoint in enumerate(image_keypoints):
                 if keypoint[0] == 0 or keypoint[1] == 0:
                     annotations["P{}x".format(num)] = image_keypoints[0][0]
                     annotations["P{}y".format(num)] = image_keypoints[0][1]
                     annotations["P{}occluded".format(num)] = True
                 else:
                     annotations["P{}x".format(num+1)] = keypoint[0]
                     annotations["P{}y".format(num+1)] = keypoint[1]
         else:
             raise ValueError("This script only generates 10 & 80 keypoints dataset.")

convert dataset command

python data_utils.py --afw_data_path /home/nxin/zyan/cv_samples_v1.4.0/fpenet/afw --output_json_path /home/nxin/zyan/cv_samples_v1.4.0/fpenet/data/afw/afw.json --afw_image_save_path /workspace/tao-experiments/fpenet/afw --num_key_points 27

train error

2022-10-24 18:02:41,861 [INFO] root: Registry: ['nvcr.io']
2022-10-24 18:02:41,904 [INFO] tlt.components.instance_handler.local_instance: Running command in container: nvcr.io/nvidia/tao/tao-toolkit-tf:v3.22.05-tf1.15.5-py3
2022-10-24 18:02:41,982 [WARNING] tlt.components.docker_handler.docker_handler: 
Docker will run the commands as root. If you would like to retain your
local host permissions, please add the "user":"UID:GID" in the
DockerOptions portion of the "/home/nxin/.tao_mounts.json" file. You can obtain your
users UID and GID by using the "id -u" and "id -g" commands on the
terminal.
2022-10-24 10:02:42.603162: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
WARNING:tensorflow:Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
/usr/local/lib/python3.6/dist-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.5) or chardet (3.0.4) doesn't match a supported version!
  RequestsDependencyWarning)
Using TensorFlow backend.
WARNING:tensorflow:From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/checkpoint_saver_hook.py:25: The name tf.train.CheckpointSaverHook is deprecated. Please use tf.estimator.CheckpointSaverHook instead.

2022-10-24 10:02:44,431 [WARNING] tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/checkpoint_saver_hook.py:25: The name tf.train.CheckpointSaverHook is deprecated. Please use tf.estimator.CheckpointSaverHook instead.

WARNING 2022-10-24 10:02:44,960| tensorflow: Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
/usr/local/lib/python3.6/dist-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.5) or chardet (3.0.4) doesn't match a supported version!
  RequestsDependencyWarning)
Using TensorFlow backend.
WARNING 2022-10-24 10:02:46,637| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/checkpoint_saver_hook.py:25: The name tf.train.CheckpointSaverHook is deprecated. Please use tf.estimator.CheckpointSaverHook instead.

WARNING 2022-10-24 10:02:46,647| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py:95: The name tf.logging.set_verbosity is deprecated. Please use tf.compat.v1.logging.set_verbosity instead.

WARNING 2022-10-24 10:02:46,647| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py:95: The name tf.logging.INFO is deprecated. Please use tf.compat.v1.logging.INFO instead.

/usr/local/lib/python3.6/dist-packages/driveix/fpenet/scripts/train.py:105: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
/workspace/tao-experiments/fpenet/models/exp1
WARNING 2022-10-24 10:02:46,984| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/dataloader/fpenet_dataloader.py:249: The name tf.FixedLenFeature is deprecated. Please use tf.io.FixedLenFeature instead.

WARNING 2022-10-24 10:02:46,988| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:153: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.

/workspace/tao-experiments/fpenet/data/tfrecords/afw/FpeTfRecords/data.tfrecords: 50
Total Samples: 50
WARNING 2022-10-24 10:02:47,030| tensorflow: From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/autograph/converters/directives.py:119: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.

WARNING 2022-10-24 10:02:47,061| tensorflow: Entity <bound method Processor.__call__ of <modulus.processors.parse_example_proto.ParseExampleProto object at 0x7f17e0a50b00>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method Processor.__call__ of <modulus.processors.parse_example_proto.ParseExampleProto object at 0x7f17e0a50b00>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING 2022-10-24 10:02:47,074| tensorflow: Entity <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f17f404bb70>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f17f404bb70>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING 2022-10-24 10:02:48,456| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/dataloader/fpenet_dataloader.py:562: The name tf.matrix_inverse is deprecated. Please use tf.linalg.inv instead.

WARNING 2022-10-24 10:02:48,464| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,468| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,469| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/common/augment/random_gamma.py:75: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.

WARNING 2022-10-24 10:02:48,475| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,636| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,640| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,647| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,810| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,813| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,820| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,989| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,992| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:48,999| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,165| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,169| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,176| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,340| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,344| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,351| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,514| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,518| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,525| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,744| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,748| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,755| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,918| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,921| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:49,928| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,090| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,094| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,101| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,263| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,267| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,274| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,436| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,440| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,446| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,609| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,612| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,619| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,780| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,784| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,791| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,955| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,959| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:50,966| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,128| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,131| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,138| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,302| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,305| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,312| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,475| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,479| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,485| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,648| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,652| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,659| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,888| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,891| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:51,898| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,060| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,063| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,070| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,235| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,239| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,246| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,409| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,412| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,419| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,582| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,585| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,592| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,754| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,757| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,764| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,927| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,930| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:52,937| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,099| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,103| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,110| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,272| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,276| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,283| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,445| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,449| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,456| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,621| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,625| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,632| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,795| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,798| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,805| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,967| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,970| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:53,977| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:02:54,014| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/visualization/fpenet_visualization.py:62: The name tf.summary.image is deprecated. Please use tf.compat.v1.summary.image instead.

INFO    2022-10-24 10:02:54,015| root: model type is: FpeNet_public
WARNING 2022-10-24 10:02:54,025| tensorflow: From /opt/nvidia/third_party/keras/tensorflow_backend.py:183: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.

WARNING 2022-10-24 10:02:54,279| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:517: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.

WARNING 2022-10-24 10:02:54,481| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:174: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.

WARNING 2022-10-24 10:02:54,481| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:190: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.

WARNING 2022-10-24 10:02:54,481| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:199: The name tf.is_variable_initialized is deprecated. Please use tf.compat.v1.is_variable_initialized instead.

WARNING 2022-10-24 10:02:55,211| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:206: The name tf.variables_initializer is deprecated. Please use tf.compat.v1.variables_initializer instead.

/usr/local/lib/python3.6/dist-packages/keras/engine/saving.py:292: UserWarning: No training configuration found in save file: the model was *not* compiled. Compile it manually.
  warnings.warn('No training configuration found in save file: '
INFO    2022-10-24 10:02:56,387| driveix.fpenet.models.fpenet_basemodel: Loading weights from pretrained model file. /workspace/tao-experiments/fpenet/pretrained_models/fpenet_vtrainable_v1.0/model.tlt
WARNING 2022-10-24 10:03:19,593| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,597| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,606| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,609| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,618| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,622| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,632| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,636| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,646| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,650| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,659| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,663| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,672| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,676| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,686| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,690| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,700| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,703| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,713| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,716| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,725| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,729| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,739| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,742| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,751| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,755| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,764| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,768| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,777| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,781| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,790| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,794| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,803| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,807| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,816| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,820| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,829| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,833| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,842| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,846| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,855| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,859| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,868| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,872| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,881| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,885| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,894| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,898| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,907| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,910| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,920| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,923| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,933| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,936| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,946| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,950| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,959| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,962| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,971| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,975| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,984| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,988| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:19,997| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-24 10:03:20,001| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
/workspace/tao-experiments/fpenet/data/tfrecords/afw/FpeTfRecords/data.tfrecords: 50
Total Samples: 50
WARNING 2022-10-24 10:03:21,017| tensorflow: Entity <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f17f404bb70>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f17f404bb70>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
INFO    2022-10-24 10:03:22,520| __main__: Build trainer finished. Starting training...
WARNING 2022-10-24 10:03:22,525| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/trainers/fpenet_trainer.py:239: The name tf.train.get_global_step is deprecated. Please use tf.compat.v1.train.get_global_step instead.

WARNING 2022-10-24 10:03:22,527| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:61: The name tf.train.LoggingTensorHook is deprecated. Please use tf.estimator.LoggingTensorHook instead.

WARNING 2022-10-24 10:03:22,528| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:62: The name tf.train.StopAtStepHook is deprecated. Please use tf.estimator.StopAtStepHook instead.

WARNING 2022-10-24 10:03:22,528| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:75: The name tf.train.StepCounterHook is deprecated. Please use tf.estimator.StepCounterHook instead.

INFO    2022-10-24 10:03:22,528| tensorflow: Create CheckpointSaverHook.
WARNING 2022-10-24 10:03:22,528| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:104: The name tf.train.SummarySaverHook is deprecated. Please use tf.estimator.SummarySaverHook instead.

INFO    2022-10-24 10:03:26,869| tensorflow: Graph was finalized.
INFO    2022-10-24 10:03:28,318| tensorflow: Running local_init_op.
INFO    2022-10-24 10:03:29,808| tensorflow: Done running local_init_op.
INFO    2022-10-24 10:03:40,529| tensorflow: Saving checkpoints for step-0.
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1365, in _do_call
    return fn(*args)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1350, in _run_fn
    target_list, run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1443, in _call_tf_sessionrun
    run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: 2 root error(s) found.
  (0) Invalid argument: {{function_node __inference_Dataset_map__map_func_set_random_wrapper_38375}} assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]
	 [[RegexFullMatch_55/_1508]]
  (1) Invalid argument: {{function_node __inference_Dataset_map__map_func_set_random_wrapper_38375}} assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]
0 successful operations.
0 derived errors ignored.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py", line 141, in <module>
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py", line 137, in main
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/trainers/fpenet_trainer.py", line 286, in train
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/trainers/fpenet_trainer.py", line 332, in run_training_loop
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 754, in run
    run_metadata=run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1360, in run
    raise six.reraise(*original_exc_info)
  File "/usr/local/lib/python3.6/dist-packages/six.py", line 696, in reraise
    raise value
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1345, in run
    return self._sess.run(*args, **kwargs)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1418, in run
    run_metadata=run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1176, in run
    return self._sess.run(*args, **kwargs)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 956, in run
    run_metadata_ptr)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1180, in _run
    feed_dict_tensor, options, run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1359, in _do_run
    run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1384, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: 2 root error(s) found.
  (0) Invalid argument:  assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]
	 [[RegexFullMatch_55/_1508]]
  (1) Invalid argument:  assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]
0 successful operations.
0 derived errors ignored.
Traceback (most recent call last):
  File "/usr/local/bin/fpenet", line 8, in <module>
    sys.exit(main())
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/entrypoint/fpenet.py", line 12, in main
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/common/entrypoint/entrypoint.py", line 300, in launch_job
AssertionError: Process run failed.
2022-10-24 18:04:03,294 [INFO] tlt.components.docker_handler.docker_handler: Stopping container.

How should I train custom keypoints with fpenet?

Could you check the label ?

I didn’t notice the problem with the label.
How the labels affect the work of the fpenet and how the custom keypoints should be modified?
My json file:
afw.json (12.8 MB)

I have read the documentation:
emotionnet-fpenet-gazenet-json-label-data-format
But I don’t quite understand.I know that the top-down keypoint algorithm needs to get the bbox first.But I don’t see that the sample data provides bbox, how the sample data bbox is obtained?
And I don’t know the difference between the face_outer_bbox and face_tight_bbox.Isn’t there supposed to be only one bbox?How should I get them in custom dataset?
My annotations file only has one bbox, how should I assign values to both of them?

afw.json (16.1 MB)
add FaceBbox, assign the same value of face_outer_bbox and face_tight_bbox
still something wrong, train error:

2022-10-25 12:34:05,506 [INFO] root: Registry: ['nvcr.io']
2022-10-25 12:34:05,547 [INFO] tlt.components.instance_handler.local_instance: Running command in container: nvcr.io/nvidia/tao/tao-toolkit-tf:v3.22.05-tf1.15.5-py3
2022-10-25 12:34:05,625 [WARNING] tlt.components.docker_handler.docker_handler: 
Docker will run the commands as root. If you would like to retain your
local host permissions, please add the "user":"UID:GID" in the
DockerOptions portion of the "/home/nxin/.tao_mounts.json" file. You can obtain your
users UID and GID by using the "id -u" and "id -g" commands on the
terminal.
2022-10-25 04:34:06.197949: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
WARNING:tensorflow:Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
/usr/local/lib/python3.6/dist-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.5) or chardet (3.0.4) doesn't match a supported version!
  RequestsDependencyWarning)
Using TensorFlow backend.
WARNING:tensorflow:From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/checkpoint_saver_hook.py:25: The name tf.train.CheckpointSaverHook is deprecated. Please use tf.estimator.CheckpointSaverHook instead.

2022-10-25 04:34:08,035 [WARNING] tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/checkpoint_saver_hook.py:25: The name tf.train.CheckpointSaverHook is deprecated. Please use tf.estimator.CheckpointSaverHook instead.

WARNING 2022-10-25 04:34:08,559| tensorflow: Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
/usr/local/lib/python3.6/dist-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.5) or chardet (3.0.4) doesn't match a supported version!
  RequestsDependencyWarning)
Using TensorFlow backend.
WARNING 2022-10-25 04:34:10,099| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/checkpoint_saver_hook.py:25: The name tf.train.CheckpointSaverHook is deprecated. Please use tf.estimator.CheckpointSaverHook instead.

WARNING 2022-10-25 04:34:10,107| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py:95: The name tf.logging.set_verbosity is deprecated. Please use tf.compat.v1.logging.set_verbosity instead.

WARNING 2022-10-25 04:34:10,108| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py:95: The name tf.logging.INFO is deprecated. Please use tf.compat.v1.logging.INFO instead.

/usr/local/lib/python3.6/dist-packages/driveix/fpenet/scripts/train.py:105: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
/workspace/tao-experiments/fpenet/models/exp1
WARNING 2022-10-25 04:34:10,440| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/dataloader/fpenet_dataloader.py:249: The name tf.FixedLenFeature is deprecated. Please use tf.io.FixedLenFeature instead.

WARNING 2022-10-25 04:34:10,444| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:153: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.

/workspace/tao-experiments/fpenet/data/tfrecords/afw/FpeTfRecords/data.tfrecords: 50
Total Samples: 50
WARNING 2022-10-25 04:34:10,486| tensorflow: From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/autograph/converters/directives.py:119: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.

WARNING 2022-10-25 04:34:10,515| tensorflow: Entity <bound method Processor.__call__ of <modulus.processors.parse_example_proto.ParseExampleProto object at 0x7f2490fb3b70>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method Processor.__call__ of <modulus.processors.parse_example_proto.ParseExampleProto object at 0x7f2490fb3b70>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING 2022-10-25 04:34:10,527| tensorflow: Entity <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f24a4587be0>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f24a4587be0>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING 2022-10-25 04:34:11,900| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/dataloader/fpenet_dataloader.py:562: The name tf.matrix_inverse is deprecated. Please use tf.linalg.inv instead.

WARNING 2022-10-25 04:34:11,908| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:11,912| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:11,913| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/common/augment/random_gamma.py:75: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.

WARNING 2022-10-25 04:34:11,919| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,081| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,085| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,091| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,252| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,256| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,263| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,424| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,427| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,435| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,595| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,599| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,606| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,767| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,771| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,777| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,940| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,944| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:12,951| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,171| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,175| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,181| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,343| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,346| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,353| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,513| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,517| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,524| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,684| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,687| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,694| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,856| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,859| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:13,866| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,026| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,030| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,037| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,198| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,202| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,209| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,371| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,375| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,382| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,543| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,547| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,554| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,715| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,719| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,726| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,888| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,892| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:14,898| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,062| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,066| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,073| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,301| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,304| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,311| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,473| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,476| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,483| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,649| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,652| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,659| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,820| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,824| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,831| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,993| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:15,997| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,004| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,164| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,168| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,175| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,336| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,340| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,347| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,508| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,511| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,519| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,681| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,685| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,691| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,853| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,856| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:16,863| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,028| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,032| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,039| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,201| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,204| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,211| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,373| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,377| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,383| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:17,418| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/visualization/fpenet_visualization.py:62: The name tf.summary.image is deprecated. Please use tf.compat.v1.summary.image instead.

INFO    2022-10-25 04:34:17,420| root: model type is: FpeNet_public
WARNING 2022-10-25 04:34:17,430| tensorflow: From /opt/nvidia/third_party/keras/tensorflow_backend.py:183: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.

WARNING 2022-10-25 04:34:17,682| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:517: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.

WARNING 2022-10-25 04:34:17,888| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:174: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.

WARNING 2022-10-25 04:34:17,888| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:190: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.

WARNING 2022-10-25 04:34:17,888| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:199: The name tf.is_variable_initialized is deprecated. Please use tf.compat.v1.is_variable_initialized instead.

WARNING 2022-10-25 04:34:18,598| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:206: The name tf.variables_initializer is deprecated. Please use tf.compat.v1.variables_initializer instead.

/usr/local/lib/python3.6/dist-packages/keras/engine/saving.py:292: UserWarning: No training configuration found in save file: the model was *not* compiled. Compile it manually.
  warnings.warn('No training configuration found in save file: '
INFO    2022-10-25 04:34:19,731| driveix.fpenet.models.fpenet_basemodel: Loading weights from pretrained model file. /workspace/tao-experiments/fpenet/pretrained_models/fpenet_vtrainable_v1.0/model.tlt
WARNING 2022-10-25 04:34:41,817| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,820| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,829| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,833| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,842| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,846| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,855| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,859| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,868| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,872| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,880| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,884| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,893| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,897| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,905| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,909| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,918| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,921| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,930| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,934| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,943| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,946| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,955| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,959| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,968| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,971| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,980| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,984| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,993| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:41,996| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,005| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,009| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,018| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,021| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,030| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,034| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,043| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,047| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,056| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,059| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,068| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,072| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,081| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,085| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,094| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,097| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,106| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,110| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,119| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,122| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,131| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,135| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,144| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,147| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,156| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,160| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,169| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,173| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,182| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,185| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,194| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,198| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,207| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 04:34:42,210| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
/workspace/tao-experiments/fpenet/data/tfrecords/afw/FpeTfRecords/data.tfrecords: 50
Total Samples: 50
WARNING 2022-10-25 04:34:43,184| tensorflow: Entity <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f24a4587be0>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f24a4587be0>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
INFO    2022-10-25 04:34:44,615| __main__: Build trainer finished. Starting training...
WARNING 2022-10-25 04:34:44,620| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/trainers/fpenet_trainer.py:239: The name tf.train.get_global_step is deprecated. Please use tf.compat.v1.train.get_global_step instead.

WARNING 2022-10-25 04:34:44,622| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:61: The name tf.train.LoggingTensorHook is deprecated. Please use tf.estimator.LoggingTensorHook instead.

WARNING 2022-10-25 04:34:44,622| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:62: The name tf.train.StopAtStepHook is deprecated. Please use tf.estimator.StopAtStepHook instead.

WARNING 2022-10-25 04:34:44,622| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:75: The name tf.train.StepCounterHook is deprecated. Please use tf.estimator.StepCounterHook instead.

INFO    2022-10-25 04:34:44,622| tensorflow: Create CheckpointSaverHook.
WARNING 2022-10-25 04:34:44,622| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:104: The name tf.train.SummarySaverHook is deprecated. Please use tf.estimator.SummarySaverHook instead.

INFO    2022-10-25 04:34:48,893| tensorflow: Graph was finalized.
INFO    2022-10-25 04:34:50,324| tensorflow: Running local_init_op.
INFO    2022-10-25 04:34:51,786| tensorflow: Done running local_init_op.
INFO    2022-10-25 04:35:02,371| tensorflow: Saving checkpoints for step-0.
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1365, in _do_call
    return fn(*args)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1350, in _run_fn
    target_list, run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1443, in _call_tf_sessionrun
    run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: 2 root error(s) found.
  (0) Invalid argument: {{function_node __inference_Dataset_map__map_func_set_random_wrapper_38375}} assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]
	 [[RegexFullMatch_55/_1508]]
  (1) Invalid argument: {{function_node __inference_Dataset_map__map_func_set_random_wrapper_38375}} assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]
0 successful operations.
0 derived errors ignored.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py", line 141, in <module>
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py", line 137, in main
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/trainers/fpenet_trainer.py", line 286, in train
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/trainers/fpenet_trainer.py", line 332, in run_training_loop
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 754, in run
    run_metadata=run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1360, in run
    raise six.reraise(*original_exc_info)
  File "/usr/local/lib/python3.6/dist-packages/six.py", line 696, in reraise
    raise value
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1345, in run
    return self._sess.run(*args, **kwargs)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1418, in run
    run_metadata=run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1176, in run
    return self._sess.run(*args, **kwargs)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 956, in run
    run_metadata_ptr)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1180, in _run
    feed_dict_tensor, options, run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1359, in _do_run
    run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1384, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: 2 root error(s) found.
  (0) Invalid argument:  assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]
	 [[RegexFullMatch_55/_1508]]
  (1) Invalid argument:  assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]
0 successful operations.
0 derived errors ignored.
Traceback (most recent call last):
  File "/usr/local/bin/fpenet", line 8, in <module>
    sys.exit(main())
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/entrypoint/fpenet.py", line 12, in main
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/common/entrypoint/entrypoint.py", line 300, in launch_job
AssertionError: Process run failed.
2022-10-25 12:35:24,821 [INFO] tlt.components.docker_handler.docker_handler: Stopping container.

From the afw.json file, I find that

Some are 0.0.

You can try to set to non zero.

More, you can download the fpenet notebook to get familiar with the fpenet-generic.
https://docs.nvidia.com/metropolis/TLT/tlt-user-guide/text/tlt_quick_start_guide.html#download-jupyter-notebook

My bad, index number should +1 when convert json.I have wrong index before.
I modified the coordinate greater than 0 too.

modified:

I think the index should be right now.
afw.json (17.9 MB)
train flie:

__class_name__: FpeNetTrainer
checkpoint_dir: /workspace/tao-experiments/fpenet/models/exp1
checkpoint_n_epoch: 1
dataloader:
  __class_name__: FpeNetDataloader
  augmentation_info:
    augmentation_resize_probability: 0.5
    augmentation_resize_scale: 1.6
    enable_occlusion_augmentation: true
    enable_online_augmentation: true
    enable_resize_augmentation: true
    gamma_augmentation:
      gamma_max: 1.6
      gamma_min: 0.6
      gamma_probability: 0.1
      gamma_type: uniform
    modulus_spatial_augmentation:
      hflip_probability: 0.0
      rotate_rad_max: 0.35
      translate_max_x: 10
      translate_max_y: 10
      zoom_max: 1.2
      zoom_min: 0.8
    patch_probability: 0.5
    size_to_image_ratio: 0.5
    mask_augmentation_patch: true
  batch_size: 32
  dataset_info:
    image_extension: png
    no_occlusion_masking_sets: none
    root_path:
    tfrecord_folder_name: FpeTfRecords
    tfrecords_directory_path: /workspace/tao-experiments/fpenet/data/tfrecords
    tfrecords_set_id_train: afw
    tfrecords_set_id_val: afw
    tfrecord_file_name: data.tfrecords
    use_extra_dataset: false
  image_info:
    image:
      channel: 1
      height: 80
      width: 80
  kpiset_info:
    tfrecords_set_id_kpi: afw
  num_keypoints: 27
enable_visualization: true
log_every_n_secs: 10
loss:
  __class_name__: FpeLoss
  kpts_coeff: 0.01
  loss_type: square_euclidean
  mask_occ: true
  weights_dict: null
  elt_loss_info:
    elt_alpha: 0.5
    enable_elt_loss: true
    modulus_spatial_augmentation:
      hflip_probability: 0.0
      rotate_rad_max: 0.35
      translate_max_x: 10
      translate_max_y: 10
      zoom_max: 1.2
      zoom_min: 0.8
model:
  __class_name__: FpeNetBaseModel
  model_parameters:
    beta: 0.1
    pretrained_model_path: /workspace/tao-experiments/fpenet/pretrained_models/fpenet_vtrainable_v1.0/model.tlt
    regularizer_type: l2
    regularizer_weight: 1.0e-05
    type: FpeNet_public
num_epoch: 50
num_keypoints: 27
optimizer:
  __class_name__: AdamOptimizer
  beta1: 0.9
  beta2: 0.999
  epsilon: 1.0e-08
  learning_rate_schedule:
    __class_name__: SoftstartAnnealingLearningRateSchedule
    annealing: 0.5
    base_learning_rate: 0.002
    last_step: 1000000
    min_learning_rate: 1.0e-04
    soft_start: 0.3
random_seed: 35
visualize_num_images: 3

tfrecords:
data.tfrecords (1.8 MB)

another error:

2022-10-25 18:30:14,641 [INFO] root: Registry: ['nvcr.io']
2022-10-25 18:30:14,682 [INFO] tlt.components.instance_handler.local_instance: Running command in container: nvcr.io/nvidia/tao/tao-toolkit-tf:v3.22.05-tf1.15.5-py3
2022-10-25 18:30:14,759 [WARNING] tlt.components.docker_handler.docker_handler: 
Docker will run the commands as root. If you would like to retain your
local host permissions, please add the "user":"UID:GID" in the
DockerOptions portion of the "/home/nxin/.tao_mounts.json" file. You can obtain your
users UID and GID by using the "id -u" and "id -g" commands on the
terminal.
2022-10-25 10:30:15.362791: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
WARNING:tensorflow:Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
/usr/local/lib/python3.6/dist-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.5) or chardet (3.0.4) doesn't match a supported version!
  RequestsDependencyWarning)
Using TensorFlow backend.
WARNING:tensorflow:From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/checkpoint_saver_hook.py:25: The name tf.train.CheckpointSaverHook is deprecated. Please use tf.estimator.CheckpointSaverHook instead.

2022-10-25 10:30:17,122 [WARNING] tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/checkpoint_saver_hook.py:25: The name tf.train.CheckpointSaverHook is deprecated. Please use tf.estimator.CheckpointSaverHook instead.

WARNING 2022-10-25 10:30:17,640| tensorflow: Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
/usr/local/lib/python3.6/dist-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.5) or chardet (3.0.4) doesn't match a supported version!
  RequestsDependencyWarning)
Using TensorFlow backend.
WARNING 2022-10-25 10:30:19,184| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/checkpoint_saver_hook.py:25: The name tf.train.CheckpointSaverHook is deprecated. Please use tf.estimator.CheckpointSaverHook instead.

WARNING 2022-10-25 10:30:19,193| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py:95: The name tf.logging.set_verbosity is deprecated. Please use tf.compat.v1.logging.set_verbosity instead.

WARNING 2022-10-25 10:30:19,193| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py:95: The name tf.logging.INFO is deprecated. Please use tf.compat.v1.logging.INFO instead.

/usr/local/lib/python3.6/dist-packages/driveix/fpenet/scripts/train.py:105: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
/workspace/tao-experiments/fpenet/models/exp1
WARNING 2022-10-25 10:30:19,529| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/dataloader/fpenet_dataloader.py:249: The name tf.FixedLenFeature is deprecated. Please use tf.io.FixedLenFeature instead.

WARNING 2022-10-25 10:30:19,533| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:153: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.

/workspace/tao-experiments/fpenet/data/tfrecords/afw/FpeTfRecords/data.tfrecords: 3178
Total Samples: 3178
WARNING 2022-10-25 10:30:19,577| tensorflow: From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/autograph/converters/directives.py:119: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.

WARNING 2022-10-25 10:30:19,606| tensorflow: Entity <bound method Processor.__call__ of <modulus.processors.parse_example_proto.ParseExampleProto object at 0x7f31b6778b00>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method Processor.__call__ of <modulus.processors.parse_example_proto.ParseExampleProto object at 0x7f31b6778b00>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING 2022-10-25 10:30:19,617| tensorflow: Entity <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f31c9d74b70>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f31c9d74b70>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
WARNING 2022-10-25 10:30:20,987| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/dataloader/fpenet_dataloader.py:562: The name tf.matrix_inverse is deprecated. Please use tf.linalg.inv instead.

WARNING 2022-10-25 10:30:20,996| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:20,999| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,001| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/common/augment/random_gamma.py:75: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.

WARNING 2022-10-25 10:30:21,006| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,165| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,169| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,176| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,334| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,338| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,344| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,504| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,508| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,514| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,673| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,676| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,683| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,843| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,846| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:21,853| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,014| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,017| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,024| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,242| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,246| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,253| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,411| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,415| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,421| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,580| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,583| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,590| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,748| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,752| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,759| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,918| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,921| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:22,928| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,088| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,092| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,098| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,259| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,262| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,269| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,429| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,433| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,440| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,600| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,604| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,610| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,770| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,773| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,780| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,941| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,945| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:23,952| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,112| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,115| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,123| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,349| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,353| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,360| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,518| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,522| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,528| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,689| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,692| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,699| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,858| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,862| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:24,868| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,027| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,030| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,037| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,200| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,203| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,210| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,370| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,374| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,381| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,539| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,543| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,550| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,709| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,713| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,719| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,878| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,882| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:25,889| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,052| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,055| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,063| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,224| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,227| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,234| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,393| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,396| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,403| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:26,438| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/visualization/fpenet_visualization.py:62: The name tf.summary.image is deprecated. Please use tf.compat.v1.summary.image instead.

INFO    2022-10-25 10:30:26,440| root: model type is: FpeNet_public
WARNING 2022-10-25 10:30:26,450| tensorflow: From /opt/nvidia/third_party/keras/tensorflow_backend.py:183: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.

WARNING 2022-10-25 10:30:26,698| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:517: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.

WARNING 2022-10-25 10:30:26,896| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:174: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.

WARNING 2022-10-25 10:30:26,896| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:190: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.

WARNING 2022-10-25 10:30:26,896| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:199: The name tf.is_variable_initialized is deprecated. Please use tf.compat.v1.is_variable_initialized instead.

WARNING 2022-10-25 10:30:27,605| tensorflow: From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:206: The name tf.variables_initializer is deprecated. Please use tf.compat.v1.variables_initializer instead.

/usr/local/lib/python3.6/dist-packages/keras/engine/saving.py:292: UserWarning: No training configuration found in save file: the model was *not* compiled. Compile it manually.
  warnings.warn('No training configuration found in save file: '
INFO    2022-10-25 10:30:28,759| driveix.fpenet.models.fpenet_basemodel: Loading weights from pretrained model file. /workspace/tao-experiments/fpenet/pretrained_models/fpenet_vtrainable_v1.0/model.tlt
WARNING 2022-10-25 10:30:51,492| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,496| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,505| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,509| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,518| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,522| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,531| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,534| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,543| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,547| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,556| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,559| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,568| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,572| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,580| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,584| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,593| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,596| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,605| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,609| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,618| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,621| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,630| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,633| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,642| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,646| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,655| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,658| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,667| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,671| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,680| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,683| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,692| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,696| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,704| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,708| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,717| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,720| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,729| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,733| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,741| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,745| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,754| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,757| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,766| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,770| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,778| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,782| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,791| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,794| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,803| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,806| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,815| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,819| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,828| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,831| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,840| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,844| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,852| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,856| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,865| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,868| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,877| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
WARNING 2022-10-25 10:30:51,881| tensorflow: The operation `tf.image.convert_image_dtype` will be skipped since the input and output dtypes are identical.
/workspace/tao-experiments/fpenet/data/tfrecords/afw/FpeTfRecords/data.tfrecords: 3178
Total Samples: 3178
WARNING 2022-10-25 10:30:52,862| tensorflow: Entity <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f31c9d74b70>> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Unable to locate the source code of <bound method FpeNetDataloader._load_and_decode of <driveix.fpenet.dataloader.fpenet_dataloader.FpeNetDataloader object at 0x7f31c9d74b70>>. Note that functions defined in certain environments, like the interactive Python shell do not expose their source code. If that is the case, you should to define them in a .py source file. If you are certain the code is graph-compatible, wrap the call using @tf.autograph.do_not_convert. Original error: could not get source code
INFO    2022-10-25 10:30:54,298| __main__: Build trainer finished. Starting training...
WARNING 2022-10-25 10:30:54,303| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/trainers/fpenet_trainer.py:239: The name tf.train.get_global_step is deprecated. Please use tf.compat.v1.train.get_global_step instead.

WARNING 2022-10-25 10:30:54,305| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:61: The name tf.train.LoggingTensorHook is deprecated. Please use tf.estimator.LoggingTensorHook instead.

WARNING 2022-10-25 10:30:54,305| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:62: The name tf.train.StopAtStepHook is deprecated. Please use tf.estimator.StopAtStepHook instead.

WARNING 2022-10-25 10:30:54,305| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:75: The name tf.train.StepCounterHook is deprecated. Please use tf.estimator.StepCounterHook instead.

INFO    2022-10-25 10:30:54,305| tensorflow: Create CheckpointSaverHook.
WARNING 2022-10-25 10:30:54,306| tensorflow: From /root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/tfhooks/utils.py:104: The name tf.train.SummarySaverHook is deprecated. Please use tf.estimator.SummarySaverHook instead.

INFO    2022-10-25 10:30:58,783| tensorflow: Graph was finalized.
INFO    2022-10-25 10:30:58,785| tensorflow: Restoring parameters from /tmp/tmp63zc6g3i/model.ckpt-0
INFO    2022-10-25 10:31:00,190| tensorflow: Running local_init_op.
INFO    2022-10-25 10:31:01,661| tensorflow: Done running local_init_op.
INFO    2022-10-25 10:31:10,395| tensorflow: Saving checkpoints for step-0.
INFO    2022-10-25 10:31:34,549| modulus.hooks.sample_counter_hook: Train Samples / sec: 2.886
INFO    2022-10-25 10:31:34,550| tensorflow: elt_loss = 99.20562, epoch = 0.0, landmarks_loss = 99.21248, step = 0, total_loss = 148.81529
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1365, in _do_call
    return fn(*args)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1350, in _run_fn
    target_list, run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1443, in _call_tf_sessionrun
    run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: {{function_node __inference_Dataset_map__map_func_set_random_wrapper_38375}} assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box_8/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py", line 141, in <module>
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/scripts/train.py", line 137, in main
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/trainers/fpenet_trainer.py", line 286, in train
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/trainers/fpenet_trainer.py", line 332, in run_training_loop
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 754, in run
    run_metadata=run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1360, in run
    raise six.reraise(*original_exc_info)
  File "/usr/local/lib/python3.6/dist-packages/six.py", line 696, in reraise
    raise value
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1345, in run
    return self._sess.run(*args, **kwargs)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/monitored_session.py", line 1426, in run
    run_metadata=run_metadata))
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/core/build_wheel.runfiles/ai_infra/moduluspy/modulus/blocks/hooks/hooks.py", line 76, in after_run
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/evaluation/fpenet_evaluator.py", line 191, in evaluate
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 956, in run
    run_metadata_ptr)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1180, in _run
    feed_dict_tensor, options, run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1359, in _do_run
    run_metadata)
  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/client/session.py", line 1384, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError:  assertion failed: [target_width must be > 0.]
	 [[{{node crop_to_bounding_box_8/Assert_2/Assert}}]]
	 [[IteratorGetNext_1]]
Traceback (most recent call last):
  File "/usr/local/bin/fpenet", line 8, in <module>
    sys.exit(main())
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/fpenet/entrypoint/fpenet.py", line 12, in main
  File "/root/.cache/bazel/_bazel_root/ed34e6d125608f91724fda23656f1726/execroot/ai_infra/bazel-out/k8-fastbuild/bin/magnet/packages/driveix/build_wheel.runfiles/ai_infra/driveix/common/entrypoint/entrypoint.py", line 300, in launch_job
AssertionError: Process run failed.
2022-10-25 18:31:36,817 [INFO] tlt.components.docker_handler.docker_handler: Stopping container.

target_width must be > 0, what does that mean?
I use fpenet in cv_samples_v1.4.0, but it only supports 10 and 80 face keypoints. It’s very difficult for me to try to use custom keypoints.

No, the notebook mentions that

# The number of keypoints can be chosen from [10, 80] for this notebook
%env NUM_KEYPOINTS=80

I see that the data processing file data_utils.py only supports 10 and 80, I try other inputs and get an error, I thought this means 10 or 80?
step convert datset to required format:

ValueError: This script only generates 10 & 80 keypoints dataset.

Yes, it can be 10 or 80 in this notebook.

But for your dataset, it can set to specific keypoints as your expected.

Did you set correct keypoint in specs/dataset_config.yaml ?

Of course, I modified the num_keypoints:

env: DATASET_CONFIG=dataset_config.yaml
sets: [afw]
gt_path: 'data'
save_path: 'data/tfrecords'
gt_root_path: '/workspace/tao-experiments/fpenet'
save_root_path: '/workspace/tao-experiments/fpenet'
image_root_path: '/workspace/tao-experiments/fpenet'
tfrecord_folder: 'FpeTfRecords'
tfrecord_name: 'data.tfrecords'
num_keypoints: 27
bbox_enlarge_ratio: 1.0

After checking , I am afraid there is still something wrong in your label.
The fpenet will find the xmin, ymin, xmax, ymax of the points .
Then calculate a bbox.

For your case, most of label contains 0.1.
That means xmin =0.1 or ymin=0.1 .
Once if the xmax or ymax is also 0.1 .
Then the width of the bbox will be 0.

It will cause error “target_width must be > 0”.

I think you are right.
I can train normally now. I have some errors in my annotation file and dummy keypoints must in the bbox.
But I don’t understand why it was designed this way. If only 1 point in bbox, width and height, both of them will be 0.
Why not let users give one tight_bbox? I think fpenet find bbox is not good.
Thank you.

Do you mean sometimes it has only 1 point(P1) as below?

            "class": "FiducialPoints",
            "P1x": 0.0,
            "P1y": 0.0,

Please filter this kind of points in the annotation file before training. Thanks.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.