Training fpenet from scratch on tao for one keypoint

Please provide the following information when requesting support.

• Hardware (Desktop/rtx 2070)
• Network Type (fpenet)
• TLT Version (nvcr.io/nvidia/tao/tao-toolkit:5.0.0-tf1.15.5)
• Training spec file(—
class_name: FpeNetTrainer
checkpoint_dir: /workspace/416x416_pictures/outputs/checkpoints
checkpoint_n_epoch: 1

dataloader:
class_name: FpeNetDataloader
augmentation_info:
augmentation_resize_probability: 0.5
augmentation_resize_scale: 1.6
enable_occlusion_augmentation: false
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: 64
dataset_info:
image_extension: png
no_occlusion_masking_sets: none
root_path: /workspace/416x416_pictures/
test_file_name: data.tfrecords
tfrecord_folder_name: FpeTfRecords
tfrecords_directory_path: /workspace/416x416_pictures/tfrecords
tfrecords_set_id_train: train
tfrecords_set_id_val: validate
tfrecord_file_name: data.tfrecords
use_extra_dataset: false
image_info:
image:
channel: 1
height: 80
width: 80
kpiset_info:
tfrecords_set_id_kpi: s578-usercalibration-incar-1
num_keypoints: 1
enable_visualization: true
hooks: null
infrequent_summary_every_n_steps: 0
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: 0.01
dropout_rate: 0.5
freeze_Convlayer: null
regularizer_type: l2
regularizer_weight: 1.0e-05
train_fpe_model: true
type: FpeNet_public
use_less_face_layers: false
use_upsampling_layer: false
visualization_parameters: null
num_epoch: 50
num_keypoints: 1
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: 0.0005
last_step: 1000000
min_learning_rate: 1.0e-07
soft_start: 0.3
random_seed: 35
visualize_num_images: 3)
• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.)

LOG:root@3c1de3eceb20:/workspace/416x416_pictures# fpenet train -e /workspace/416x416_pictures/fpenet_config.yaml
2024-12-07 21:52:29.288695: I tensorflow/stream_executor/platform/default/dso_loader.cc:50] Successfully opened dynamic library libcudart.so.12
WARNING:tensorflow:Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
Using TensorFlow backend.
WARNING:tensorflow:TensorFlow will not use sklearn by default. This improves performance in some cases. To enable sklearn export the environment variable TF_ALLOW_IOLIBS=1.
2024-12-07 21:52:30,352 [TAO Toolkit] [WARNING] tensorflow 43: TensorFlow will not use sklearn by default. This improves performance in some cases. To enable sklearn export the environment variable TF_ALLOW_IOLIBS=1.
WARNING:tensorflow:TensorFlow will not use Dask by default. This improves performance in some cases. To enable Dask export the environment variable TF_ALLOW_IOLIBS=1.
2024-12-07 21:52:30,385 [TAO Toolkit] [WARNING] tensorflow 42: TensorFlow will not use Dask by default. This improves performance in some cases. To enable Dask export the environment variable TF_ALLOW_IOLIBS=1.
WARNING:tensorflow:TensorFlow will not use Pandas by default. This improves performance in some cases. To enable Pandas export the environment variable TF_ALLOW_IOLIBS=1.
2024-12-07 21:52:30,388 [TAO Toolkit] [WARNING] tensorflow 43: TensorFlow will not use Pandas by default. This improves performance in some cases. To enable Pandas export the environment variable TF_ALLOW_IOLIBS=1.
2024-12-07 21:52:31,417 [TAO Toolkit] [WARNING] tensorflow 40: Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
Using TensorFlow backend.
2024-12-07 21:52:32,455 [TAO Toolkit] [WARNING] tensorflow 43: TensorFlow will not use sklearn by default. This improves performance in some cases. To enable sklearn export the environment variable TF_ALLOW_IOLIBS=1.
2024-12-07 21:52:32,489 [TAO Toolkit] [WARNING] tensorflow 42: TensorFlow will not use Dask by default. This improves performance in some cases. To enable Dask export the environment variable TF_ALLOW_IOLIBS=1.
2024-12-07 21:52:32,492 [TAO Toolkit] [WARNING] tensorflow 43: TensorFlow will not use Pandas by default. This improves performance in some cases. To enable Pandas export the environment variable TF_ALLOW_IOLIBS=1.
2024-12-07 21:52:32,911 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/scripts/train.py:110: The name tf.logging.set_verbosity is deprecated. Please use tf.compat.v1.logging.set_verbosity instead.

2024-12-07 21:52:32,911 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/scripts/train.py:110: The name tf.logging.INFO is deprecated. Please use tf.compat.v1.logging.INFO instead.

/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/scripts/train.py:120: YAMLLoadWarning: calling yaml.load() without Loader=… is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
spec = yaml.load(yaml_file.read())
/workspace/416x416_pictures/outputs/checkpoints
2024-12-07 21:52:33,249 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.common.logging.logging 197: Log file already exists at /workspace/416x416_pictures/outputs/checkpoints/status.json
2024-12-07 21:52:33,252 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/dataloader/fpenet_dataloader.py:262: The name tf.FixedLenFeature is deprecated. Please use tf.io.FixedLenFeature instead.

2024-12-07 21:52:33,256 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/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/416x416_pictures/tfrecords/train/FpeTfRecords/data.tfrecords: 2521
Total Samples: 2521
2024-12-07 21:52:33,333 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/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.

2024-12-07 21:52:37,567 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/dataloader/fpenet_dataloader.py:575: The name tf.matrix_inverse is deprecated. Please use tf.linalg.inv instead.

2024-12-07 21:52:37,597 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,601 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,603 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/core/augment/random_gamma.py:87: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.

2024-12-07 21:52:37,610 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,711 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,715 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,723 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,825 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,829 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,837 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,938 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,942 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:37,950 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,051 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,056 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,063 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,166 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,170 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,178 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,280 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,284 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,294 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,461 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,465 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,473 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,576 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,580 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,588 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,691 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,695 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,703 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,805 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,810 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,818 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,922 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,926 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:38,934 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,037 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,042 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,049 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,152 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,156 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,164 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,267 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,271 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,279 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,382 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,386 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,394 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,496 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,500 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,508 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,610 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,614 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,623 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,725 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,729 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,736 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,839 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,844 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,851 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,955 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,959 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:39,967 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,069 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,073 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,081 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,183 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,187 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,195 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,296 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,300 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,309 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,412 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,416 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,424 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,609 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,613 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,621 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,725 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,729 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,737 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,840 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,845 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,853 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,957 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,961 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:40,969 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,073 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,077 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,085 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,189 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,193 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,201 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,305 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,309 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,317 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,419 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,424 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,431 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,535 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,539 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,547 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,650 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,654 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,662 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,766 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,770 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,778 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,882 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,886 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,894 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:41,996 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,000 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,008 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,111 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,115 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,123 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,226 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,230 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,238 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,353 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,358 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,367 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,494 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,499 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,509 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,634 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,638 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,646 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,751 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,755 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,763 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,867 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,872 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,879 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,991 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:42,995 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,003 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,110 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,115 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,122 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,326 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,330 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,338 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,443 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,447 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,455 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,562 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,566 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,574 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,680 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,684 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,692 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,795 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,799 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,806 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,910 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,914 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:43,922 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,024 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,028 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,036 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,138 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,142 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,150 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,253 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,257 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,265 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,369 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,373 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,380 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,483 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,487 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,495 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,598 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,602 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,609 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,712 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,716 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,723 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,828 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,832 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,840 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,944 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,948 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:44,955 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,059 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,063 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,071 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,173 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,177 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,185 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,241 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/visualization/fpenet_visualization.py:74: The name tf.summary.image is deprecated. Please use tf.compat.v1.summary.image instead.

2024-12-07 21:52:45,243 [TAO Toolkit] [INFO] root 2082: model type is: FpeNet_public
2024-12-07 21:52:45,257 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/third_party/keras/tensorflow_backend.py:195: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.

2024-12-07 21:52:45,496 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.fpenet.models.fpenet_basemodel 166: This model will be trained from scratch.
2024-12-07 21:52:45,608 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,612 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,622 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,626 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,637 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,641 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,652 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,656 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,666 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,670 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,680 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,684 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,694 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,698 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,709 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,713 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,723 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,727 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,738 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,742 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,752 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,756 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,767 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,771 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,782 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,786 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,796 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,800 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,810 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,815 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,825 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,829 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,839 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,843 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,854 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,858 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,868 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,872 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,882 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,886 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,897 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,901 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,911 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,915 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,926 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,930 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,940 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,944 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,955 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,959 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,969 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,973 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,984 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,988 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:45,998 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,002 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,013 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,017 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,027 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,031 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,042 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,046 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,056 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,060 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,071 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,075 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,085 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,089 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,100 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,104 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,114 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,119 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,129 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,133 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,143 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,147 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,157 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,162 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,172 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,176 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,186 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,190 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,201 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,205 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,215 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,219 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,229 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,234 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,244 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,248 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,258 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,263 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,273 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,277 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,288 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,292 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,302 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,306 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,317 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,321 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,331 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,335 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,346 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,350 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,360 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,364 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,375 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,379 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,389 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,393 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,404 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,408 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,418 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,422 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,433 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,437 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,558 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,562 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,572 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,576 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,586 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,591 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,601 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,605 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,615 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,619 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,629 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:46,634 [TAO Toolkit] [WARNING] tensorflow 1776: The operation tf.image.convert_image_dtype will be skipped since the input and output dtypes are identical.
2024-12-07 21:52:47,870 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.fpenet.trainers.fpenet_trainer 217: Evaluation model file path wasn’t provided. Getting the latest checkpoint in /workspace/416x416_pictures/outputs/checkpoints
Checkpoint results dir /workspace/416x416_pictures/outputs/checkpoints
2024-12-07 21:52:47,871 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.fpenet.trainers.fpenet_trainer 227: Evaluating using the model at None
/workspace/416x416_pictures/tfrecords/validate/FpeTfRecords/data.tfrecords: 348
Total Samples: 348
2024-12-07 21:52:51,305 [TAO Toolkit] [INFO] main 157: Build trainer finished. Starting training…
2024-12-07 21:52:51,305 [TAO Toolkit] [INFO] root 2102: Starting fpenet training.
2024-12-07 21:52:51,314 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/trainers/fpenet_trainer.py:290: The name tf.train.get_global_step is deprecated. Please use tf.compat.v1.train.get_global_step instead.

2024-12-07 21:52:51,316 [TAO Toolkit] [INFO] tensorflow 541: Create CheckpointSaverHook.
Checkpoint results dir /workspace/416x416_pictures/outputs/checkpoints
2024-12-07 21:52:58,877 [TAO Toolkit] [INFO] tensorflow 240: Graph was finalized.
2024-12-07 21:53:01,160 [TAO Toolkit] [INFO] tensorflow 500: Running local_init_op.
2024-12-07 21:53:05,009 [TAO Toolkit] [INFO] tensorflow 502: Done running local_init_op.
2024-12-07 21:53:05,732 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/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.

2024-12-07 21:53:22,658 [TAO Toolkit] [INFO] tensorflow 81: Saving checkpoints for step-0.
2024-12-07 21:53:52,353 [TAO Toolkit] [INFO] root 2102: 2 root error(s) found.
(0) Invalid argument: Feature: train/facebbx_h (data type: int64) is required but could not be found.
[[{{node ParseExampleProto/ParseSingleExample/ParseSingleExample}}]]
[[IteratorGetNext_1]]
(1) Invalid argument: Feature: train/facebbx_h (data type: int64) is required but could not be found.
[[{{node ParseExampleProto/ParseSingleExample/ParseSingleExample}}]]
[[IteratorGetNext_1]]
[[RegexFullMatch_31/_1160]]
0 successful operations.
0 derived errors ignored.
Traceback (most recent call last):
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/client/session.py”, line 1365, in _do_call
return fn(*args)
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/client/session.py”, line 1349, in _run_fn
return self._call_tf_sessionrun(options, feed_dict, fetch_list,
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/client/session.py”, line 1441, in _call_tf_sessionrun
return tf_session.TF_SessionRun_wrapper(self._session, options, feed_dict,
tensorflow.python.framework.errors_impl.InvalidArgumentError: 2 root error(s) found.
(0) Invalid argument: {{function_node __inference_Dataset_map__map_func_set_random_wrapper_45018}} Feature: train/facebbx_h (data type: int64) is required but could not be found.
[[{{node ParseExampleProto/ParseSingleExample/ParseSingleExample}}]]
[[IteratorGetNext_1]]
(1) Invalid argument: {{function_node __inference_Dataset_map__map_func_set_random_wrapper_45018}} Feature: train/facebbx_h (data type: int64) is required but could not be found.
[[{{node ParseExampleProto/ParseSingleExample/ParseSingleExample}}]]
[[IteratorGetNext_1]]
[[RegexFullMatch_31/_1160]]
0 successful operations.
0 derived errors ignored.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/scripts/train.py”, line 195, in
raise e
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/scripts/train.py”, line 179, in
main()
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/scripts/train.py”, line 169, in main
trainer.train()
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/trainers/fpenet_trainer.py”, line 342, in train
self.run_training_loop(
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/fpenet/trainers/fpenet_trainer.py”, line 390, in run_training_loop
sess.run(train_op)
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/training/monitored_session.py”, line 750, in run
return self._sess.run(
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/training/monitored_session.py”, line 1360, in run
raise six.reraise(*original_exc_info)
File “/usr/local/lib/python3.8/dist-packages/six.py”, line 719, in reraise
raise value
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/training/monitored_session.py”, line 1345, in run
return self._sess.run(*args, **kwargs)
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/training/monitored_session.py”, line 1413, in run
outputs = _WrappedSession.run(
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/training/monitored_session.py”, line 1176, in run
return self._sess.run(*args, **kwargs)
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/client/session.py”, line 955, in run
result = self._run(None, fetches, feed_dict, options_ptr,
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/client/session.py”, line 1179, in _run
results = self._do_run(handle, final_targets, final_fetches,
File “/usr/local/lib/python3.8/dist-packages/tensorflow_core/python/client/session.py”, line 1358, in _do_run
return self._do_call(_run_fn, feeds, fetches, targets, options,
File “/usr/local/lib/python3.8/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: Feature: train/facebbx_h (data type: int64) is required but could not be found.
[[{{node ParseExampleProto/ParseSingleExample/ParseSingleExample}}]]
[[IteratorGetNext_1]]
(1) Invalid argument: Feature: train/facebbx_h (data type: int64) is required but could not be found.
[[{{node ParseExampleProto/ParseSingleExample/ParseSingleExample}}]]
[[IteratorGetNext_1]]
[[RegexFullMatch_31/_1160]]
0 successful operations.
0 derived errors ignored.
Execution status: FAIL

on 416x416 images I havea annotated “lim” bboxes and “rfx” keypoints. The rfx are points in lim … (it is analog to face=>lim ; facelandmarks=> rfx… but there is just one point as landmark) Then I cropped the boboxes and resized them to 80x80. then I created a coco json ONLY for Keypoints. Then I converted the cocoson to TFRecords… Fpenet Training on Tao docker asks for " train/facebbx" Can we not train the landmarks on cropped images (analog to cropped face images)… But on deepstream, face works as pgie, landmarks as sgie, and they are seperate models, arnt they? Can I not train one model just for one keypoint on fpenet?

May I know if you can run default notebook successfully?
For own dataset, please note that the format is mentioned in Data Annotation Format - NVIDIA Docs.

I ran trained a few models successfully in yolo train

Thanks for the answer, I will change it to to that format… but still,
the question is, how strict does it to be? Which variables must be defined in that format?
Do I have to use “class”: “FaceBbox”, “face_outer_bboxx”:class": “eyes” FiducialPoints… …do I have to define a box?? I just need one point actually" What is face_outer_bboxx and face_tight_bboxx? Can I not define just one point?? Or one bbox and one point? rename classes arbitrary?

Or is it enough to define:

[ "filename": "data/001_01_02_200_06.png",
 "class": "image",
 "annotations": [
     {
         "class": "Bbox",
                    bboxx
                    bboxy
                    bboxwidth
                   bboxxhight, }

   {"class": "FiducialPoints",
         "tool-version": "1.0",
         "P1x": xxxxxxxx,
         "P1y" : yyyyyyyy } ]]

I have had used similar format, which parts would you strictly change: My format below… then the nvidia mentioned format sequentially given here below:

“images”: [
{
“id”: 0,
“file_name”: “0.jpg”,
“width”: 120,
“height”: 120
},
“annotations”: [
{
“id”: 0,
“image_id”: 0,
“category_id”: 1,
“num_keypoints”: 1,
“keypoints”: [
62.19512195121951,
59.25925925925925,
2
],
“iscrowd”: 0
},
"
“categories”: [
{
“id”: 1,
“name”: “rfx”,
“supercategory”: “none”,
“keypoints”: [
“midpoint”
],
“skeleton”:
}
]
}

+++++++++++++++++++++++++++++++++++++++++

.
{
“filename”: “data/001_01_02_200_06.png”,
“class”: “image”,
“annotations”: [
{
“class”: “FaceBbox”,
“tool-version”: “1.0”,
“Occlusion”: 0,
“face_outer_bboxx”: 269.0082935424086,
“face_outer_bboxy”: 44.33839032556304,
“face_outer_bboxwidth”: 182.97858097042064,
“face_outer_bboxheight”: 276.28773076003836,
“face_tight_bboxx”: 269.211755426433,
“face_tight_bboxy”: 147.9049289218409,
“face_tight_bboxwidth”: 182.58110482105968,
“face_tight_bboxheight”: 172.5088694283426
},
{
“class”: “FiducialPoints”,
“tool-version”: “1.0”,
“P1x”: 304.8502837500011,
“P1y”: 217.10946645000078,
“P2x”: 311.0173699500011,
“P2y”: 237.15249660000086,
.
.
“P26occluded”: true,
“P46occluded”: true,
.
.
“P68x”: 419.5885050000024,
“P68y”: 267.6976650000015,
.
.
“P104x”: 429.6,
“P104y”: 189.5,
},
{
“class”: “eyes”
“tool-version”: “1.0”,
“l_eyex”: 389.1221901922325,
“l_eyey”: 197.94528259092206,
“r_eyex”: 633.489814294182,
“r_eyey”: 10.52527209626886,
“l_status”: “open”,
“r_status”: “occluded”,
}
]
}

Facebox is needed to set. Please follow the doc’s format. The annotation file just provide all the keypoints. FPEnet will find the xmin, ymin, xmax, ymax of the points and then calculate a square face bounding box based on the key points. And then crop bounding box from image and scale the Keypoints to target resolution(for example, 80x80).
Different keypoints is possible. You can find 10 or 80 points in tao_tutorials/notebooks/tao_launcher_starter_kit/fpenet/fpenet.ipynb at main · NVIDIA/tao_tutorials · GitHub.
Also suggest you to take a look at tao_tensorflow1_backend/nvidia_tao_tf1/cv/fpenet/dataloader/fpenet_dataloader.py at main · NVIDIA/tao_tensorflow1_backend · GitHub about the dataloader.

I want to detect one point in one object.
I have annotated data and I can confirm that the bounding box for the first object is working well (yolov4_tiny tao train).

For the one point located in that object, I have tried to annotate and train with yolov4 another object bbox. I have tried to annotate with small boxes that did not go well, then with a greater bbox, both did not work well. Yolov4 tiny is converging, but the midpoint of that bbox is not a good loss function, it cannot estimate the coordinate of the midpoint adaquately.

Should I prefer training a keypoint detector? Which tao model would you recommend for accurately detecting that one single point coordinate within the recognized object ? Can FPEnet be trained to detect both a boynding box and a single keypoint (similar yo detecting face and one facelandmark)? Or would you suggest using a different model? The target is actually just a bright round light reflex point, and can be easily detected with opencv methods, but these methods sometimes lead to errors

Similar to yolo_v4 tiny, I suggest you to continue to use object detection models from TAO. For example, DINO. Please refer to DINO - NVIDIA Docs and notebook tao_tutorials/notebooks/tao_launcher_starter_kit/dino/dino.ipynb at main · NVIDIA/tao_tutorials · GitHub.

But I need a keypoint or point detection. Can I edit in tao yolo the loss function so that it would be based only on the distance of the estimation point and nor width and height of the bbox?
Dino is not a keypoint detector, why could I use it as a keypoint detector?
Which keypoint detector would be most accurate to train in tao? My jpgs would be about 100x100 in size. Detectnet, posenet, fpenet?

Could you please share several example images?

Currently in TAO, there is fpenet and CenterPose which can detect keypoint.

The reflex point (upper right) is to be recognized with a keypoint detector. Number of Keypoints is 1. Other methods to treshold and momentum with opencv are slow and lead to errors if there are surround lights, larger reflections etc. I have annotated photos. As there is only one keypoint, there is no need to pose or skleton. Would FPEnet be suitable, if I train fpenet to find the eye (the above frame actually) and the 1 reflection keypoint in it?

For your case, I would like to suggest you to use object Detector instead of keypoint detector. Because keypoints detector usually helps to understand the structure and pose of objects. But your case has only one keypoint. The usual cases for kepoints detectors are to represent the corners of an object or joints in a human body. For example, in human pose estimation, keypoints might include locations for the head, shoulders, elbows, and knees.

Instead, you can use object detector, such as YOLO_v4 or DINO.

I have tried yolov4 ; if small bboxes are annotated areund the reflex, similar objects (white dots) are found everywhere. If I male bigger bboxes, where the middlepoint is the object, than the precision is bad, as the loss function converges to estimate loss for middlepoint + loss for width + loss for hight. Would I have any chance to edit loss function in tao? Training a yolopose in python and adating the produced onnx file to deepstream also didnt work. Should I give a one more try to use exterally trained onnx in deepstream? Any orher suggestions

You can label two kinds of classes. One class is the large middlepoint. Another class is the expected reflex point. Please try to run with YOLO_v4 and DINO.

I need to find only the coordinate of reflexpoint. I mean with the middlepoint that the middlepoint of the bigger bbox annotation falls on the reflex pointü

Would you suggest to annotate like the red bbox or the green bbox, where the midflepoin of the bbox falls on the object that I want to find

I mean to generate two classes as below. One is red area. Another is yellow area.