Tao model detectnet_v2 dataset_convert : ValueError: could not convert string to float: 'fallback"'

• Network Type (Detectnet_v2)
• TLT Version (Please run “tlt info --verbose” and share “docker_tag” here)
nvidia/tao/tao-toolkit:
5.0.0-tf2.11.0:
• Training spec file(If have, please share here)

kitti_config {
root_directory_path: “/workspace/tao-experiments/data/training”
image_dir_name: “image_2”
label_dir_name: “label_1”
image_extension: “.jpg”
partition_mode: “random”
num_partitions: 2
val_split: 20
num_shards: 10
}
image_directory_path: “/workspace/tao-experiments/data/training”

when iam running the command
tao model detectnet_v2 dataset_convert
-d $SPECS_DIR/detectnet_v2_tfrecords_kitti_trainval.txt
-o $DATA_DOWNLOAD_DIR/tfrecords/kitti_trainval/kitti_trainval
-r $USER_EXPERIMENT_DIR/

iam getting the error
Converting Tfrecords for kitti trainval dataset
2024-05-20 09:57:03,454 [TAO Toolkit] [INFO] root 160: Registry: [‘nvcr.io’]
2024-05-20 09:57:03,520 [TAO Toolkit] [INFO] nvidia_tao_cli.components.instance_handler.local_instance 360: Running command in container: nvcr.io/nvidia/tao/tao-toolkit:5.0.0-tf1.15.5
2024-05-20 09:57:03,549 [TAO Toolkit] [INFO] nvidia_tao_cli.components.docker_handler.docker_handler 301: Printing tty value True
2024-05-20 04:27:05.438111: I tensorflow/stream_executor/platform/default/dso_loader.cc:50] Successfully opened dynamic library libcudart.so.12
2024-05-20 04:27:05,766 [TAO Toolkit] [WARNING] tensorflow 40: Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
Using TensorFlow backend.
2024-05-20 04:27:08,294 [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.
2024-05-20 04:27:08,365 [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.
2024-05-20 04:27:08,376 [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.
2024-05-20 04:27:12,431 [TAO Toolkit] [INFO] matplotlib.font_manager 1633: generated new fontManager
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.
2024-05-20 04:27:13,928 [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.
2024-05-20 04:27:13,957 [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.
2024-05-20 04:27:13,960 [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.
2024-05-20 04:27:14,373 [TAO Toolkit] [INFO] root 2102: Starting Object Detection Dataset Convert.
2024-05-20 04:27:14,373 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.build_converter 87: Instantiating a kitti converter
2024-05-20 04:27:14,373 [TAO Toolkit] [INFO] root 2102: Instantiating a kitti converter
2024-05-20 04:27:14,374 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.dataset_converter_lib 71: Creating output directory /home/sysadmin/Desktop/dataset/tao-getting-started_5.3.0/notebooks/tao_launcher_starter_kit/detectnet_v2/tfrecords/kitti_trainval
2024-05-20 04:27:14,374 [TAO Toolkit] [INFO] root 2102: Generating partitions
2024-05-20 04:27:14,424 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.kitti_converter_lib 176: Num images in
Train: 23866 Val: 5966
2024-05-20 04:27:14,424 [TAO Toolkit] [INFO] root 2102: Num images in
Train: 23866 Val: 5966
2024-05-20 04:27:14,424 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.kitti_converter_lib 197: Validation data in partition 0. Hence, while choosing the validationset during training choose validation_fold 0.
2024-05-20 04:27:14,424 [TAO Toolkit] [INFO] root 2102: Validation data in partition 0. Hence, while choosing the validationset during training choose validation_fold 0.
2024-05-20 04:27:14,434 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.dataset_converter_lib 166: Writing partition 0, shard 0
2024-05-20 04:27:14,434 [TAO Toolkit] [INFO] root 2102: Writing partition 0, shard 0
WARNING:tensorflow:From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py:181: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.

2024-05-20 04:27:14,434 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py:181: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.

2024-05-20 04:27:14,582 [TAO Toolkit] [INFO] root 2102: could not convert string to float: ‘fallback"’
Traceback (most recent call last):
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/scripts/dataset_convert.py”, line 168, in
raise e
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/scripts/dataset_convert.py”, line 137, in
main()
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/scripts/dataset_convert.py”, line 132, in main
converter.convert()
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py”, line 86, in convert
object_count = self._write_partitions(partitions)
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py”, line 141, in _write_partitions
shard_object_count = self._write_shard(shard, p, s)
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py”, line 185, in _write_shard
example = self._create_example_proto(frame_id)
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/kitti_converter_lib.py”, line 223, in _create_example_proto
self._add_targets(example, frame_id)
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/kitti_converter_lib.py”, line 334, in _add_targets
labels = [[l[0].encode(“utf-8”)] + [float(x) for x in l[1:]] for l in labels]
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/kitti_converter_lib.py”, line 334, in
labels = [[l[0].encode(“utf-8”)] + [float(x) for x in l[1:]] for l in labels]
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/kitti_converter_lib.py”, line 334, in
labels = [[l[0].encode(“utf-8”)] + [float(x) for x in l[1:]] for l in labels]
ValueError: could not convert string to float: ‘fallback"’
Execution status: FAIL
2024-05-20 09:57:19,140 [TAO Toolkit] [INFO] nvidia_tao_cli.components.docker_handler.docker_handler 363: Stopping container.

my sample dataset is as the crt format in the site
3.txt (194 Bytes)
motorcycle 0.0 0 0.0 1744 417 1841 533 0.0 0.0 0.0 0.0 0.0 0.0 0.0
rider 0.0 0 0.0 1744 368 1837 502 0.0 0.0 0.0 0.0 0.0 0.0 0.0

but when iam running the above tao command iam getting an error of could not convert string to float: “fallback”

Converting Tfrecords for kitti trainval dataset
2024-05-20 09:57:03,454 [TAO Toolkit] [INFO] root 160: Registry: [‘nvcr.io’]
2024-05-20 09:57:03,520 [TAO Toolkit] [INFO] nvidia_tao_cli.components.instance_handler.local_instance 360: Running command in container: nvcr.io/nvidia/tao/tao-toolkit:5.0.0-tf1.15.5
2024-05-20 09:57:03,549 [TAO Toolkit] [INFO] nvidia_tao_cli.components.docker_handler.docker_handler 301: Printing tty value True
2024-05-20 04:27:05.438111: I tensorflow/stream_executor/platform/default/dso_loader.cc:50] Successfully opened dynamic library libcudart.so.12
2024-05-20 04:27:05,766 [TAO Toolkit] [WARNING] tensorflow 40: Deprecation warnings have been disabled. Set TF_ENABLE_DEPRECATION_WARNINGS=1 to re-enable them.
Using TensorFlow backend.
2024-05-20 04:27:08,294 [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.
2024-05-20 04:27:08,365 [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.
2024-05-20 04:27:08,376 [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.
2024-05-20 04:27:12,431 [TAO Toolkit] [INFO] matplotlib.font_manager 1633: generated new fontManager
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.
2024-05-20 04:27:13,928 [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.
2024-05-20 04:27:13,957 [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.
2024-05-20 04:27:13,960 [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.
2024-05-20 04:27:14,373 [TAO Toolkit] [INFO] root 2102: Starting Object Detection Dataset Convert.
2024-05-20 04:27:14,373 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.build_converter 87: Instantiating a kitti converter
2024-05-20 04:27:14,373 [TAO Toolkit] [INFO] root 2102: Instantiating a kitti converter
2024-05-20 04:27:14,374 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.dataset_converter_lib 71: Creating output directory /home/sysadmin/Desktop/dataset/tao-getting-started_5.3.0/notebooks/tao_launcher_starter_kit/detectnet_v2/tfrecords/kitti_trainval
2024-05-20 04:27:14,374 [TAO Toolkit] [INFO] root 2102: Generating partitions
2024-05-20 04:27:14,424 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.kitti_converter_lib 176: Num images in
Train: 23866 Val: 5966
2024-05-20 04:27:14,424 [TAO Toolkit] [INFO] root 2102: Num images in
Train: 23866 Val: 5966
2024-05-20 04:27:14,424 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.kitti_converter_lib 197: Validation data in partition 0. Hence, while choosing the validationset during training choose validation_fold 0.
2024-05-20 04:27:14,424 [TAO Toolkit] [INFO] root 2102: Validation data in partition 0. Hence, while choosing the validationset during training choose validation_fold 0.
2024-05-20 04:27:14,434 [TAO Toolkit] [INFO] nvidia_tao_tf1.cv.detectnet_v2.dataio.dataset_converter_lib 166: Writing partition 0, shard 0
2024-05-20 04:27:14,434 [TAO Toolkit] [INFO] root 2102: Writing partition 0, shard 0
WARNING:tensorflow:From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py:181: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.

2024-05-20 04:27:14,434 [TAO Toolkit] [WARNING] tensorflow 137: From /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py:181: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.

2024-05-20 04:27:14,582 [TAO Toolkit] [INFO] root 2102: could not convert string to float: ‘fallback"’
Traceback (most recent call last):
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/scripts/dataset_convert.py”, line 168, in
raise e
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/scripts/dataset_convert.py”, line 137, in
main()
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/scripts/dataset_convert.py”, line 132, in main
converter.convert()
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py”, line 86, in convert
object_count = self._write_partitions(partitions)
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py”, line 141, in _write_partitions
shard_object_count = self._write_shard(shard, p, s)
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/dataset_converter_lib.py”, line 185, in _write_shard
example = self._create_example_proto(frame_id)
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/kitti_converter_lib.py”, line 223, in _create_example_proto
self._add_targets(example, frame_id)
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/kitti_converter_lib.py”, line 334, in _add_targets
labels = [[l[0].encode(“utf-8”)] + [float(x) for x in l[1:]] for l in labels]
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/kitti_converter_lib.py”, line 334, in
labels = [[l[0].encode(“utf-8”)] + [float(x) for x in l[1:]] for l in labels]
File “/usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/kitti_converter_lib.py”, line 334, in
labels = [[l[0].encode(“utf-8”)] + [float(x) for x in l[1:]] for l in labels]
ValueError: could not convert string to float: ‘fallback"’
Execution status: FAIL
2024-05-20 09:57:19,140 [TAO Toolkit] [INFO] nvidia_tao_cli.components.docker_handler.docker_handler 363: Stopping container.

how to fix this error please help

According to the rules defined by the Python programming language, a string can be converted into a floating point datatype if it contains only numerical. If it contains anything other characters like commas, spaces, or certain other characters then we face valueerror i.e. “could not convert string to float”.
Can you check the labels?
Or, you can add some debug log in /usr/local/lib/python3.8/dist-packages/nvidia_tao_tf1/cv/detectnet_v2/dataio/kitti_converter_lib.py to check the labels.

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