Hi, I’m trying to use tlt to train DetectNet_V2
I have the data and have converted it into tfrecords. But when I try to run the tlt-train I get the following error.
ValueError: No dataset tfrecords file found at path: '/workspace/tlt-experiments/data/data_with_eval/train/tfrecords’
I have verified the the records files are present in that folder. I have attached a screenshot of the contents of the folder below.
Full log is below :
Using TensorFlow backend.
2020-08-13 10:43:50.381686: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
[[39368,1],0]: A high-performance Open MPI point-to-point messaging module
was unable to find any relevant network interfaces:
Module: OpenFabrics (openib)
Host: dce1b73ec71f
Another transport will be used instead, although this may result in
lower performance.
NOTE: You can disable this warning by setting the MCA parameter
btl_base_warn_component_unused to 0.
2020-08-13 10:43:52.428678: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2020-08-13 10:43:52.452169: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-13 10:43:52.452820: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties:
name: GeForce RTX 2080 Ti major: 7 minor: 5 memoryClockRate(GHz): 1.545
pciBusID: 0000:09:00.0
2020-08-13 10:43:52.452849: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
2020-08-13 10:43:52.452906: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0
2020-08-13 10:43:52.454247: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0
2020-08-13 10:43:52.454615: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0
2020-08-13 10:43:52.456395: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0
2020-08-13 10:43:52.457664: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0
2020-08-13 10:43:52.457721: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-08-13 10:43:52.457927: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-13 10:43:52.458647: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-13 10:43:52.459226: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: 0
2020-08-13 10:43:52.459269: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
2020-08-13 10:43:53.068882: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1159] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-08-13 10:43:53.068932: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1165] 0
2020-08-13 10:43:53.068940: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1178] 0: N
2020-08-13 10:43:53.069198: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-13 10:43:53.069670: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-13 10:43:53.070111: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:983] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-08-13 10:43:53.070512: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1304] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 9831 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2080 Ti, pci bus id: 0000:09:00.0, compute capability: 7.5)
2020-08-13 10:43:53,071 [INFO] iva.detectnet_v2.scripts.train: Loading experiment spec at detect_net_config.txt.
2020-08-13 10:43:53,073 [INFO] iva.detectnet_v2.spec_handler.spec_loader: Merging specification from detect_net_config.txt
Traceback (most recent call last):
File “/usr/local/bin/tlt-train-g1”, line 8, in
sys.exit(main())
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/common/magnet_train.py”, line 55, in main
File “”, line 2, in main
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/utilities/timer.py”, line 46, in wrapped_fn
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/scripts/train.py”, line 773, in main
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/scripts/train.py”, line 691, in run_experiment
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/scripts/train.py”, line 557, in train_gridbox
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/dataloader/build_dataloader.py”, line 264, in build_dataloader
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/dataloader/drivenet_dataloader.py”, line 384, in init
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/dataloader/drivenet_dataloader.py”, line 427, in _construct_data_sources
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/dataloader/drivenet_dataloader.py”, line 290, in init
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/core/build_wheel.runfiles/ai_infra/moduluspy/modulus/modulusobject/modulusobject.py”, line 432, in wrapper
File “/home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/core/build_wheel.runfiles/ai_infra/moduluspy/modulus/blocks/data_loaders/multi_source_loader/sources/tfrecords_data_source.py”, line 62, in init
ValueError: No dataset tfrecords file found at path: ‘/workspace/tlt-experiments/data/data_with_eval/train/tfrecords’
tlt-dataset-convert output :
Using TensorFlow backend.
2020-08-13 10:49:20,496 - iva.detectnet_v2.dataio.build_converter - INFO - Instantiating a kitti converter
2020-08-13 10:49:20,498 - iva.detectnet_v2.dataio.kitti_converter_lib - INFO - Num images in
Train: 603 Val: 98
2020-08-13 10:49:20,498 - iva.detectnet_v2.dataio.kitti_converter_lib - INFO - Validation data in partition 0. Hence, while choosing the validationset during training choose validation_fold 0.
2020-08-13 10:49:20,498 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 0
WARNING:tensorflow:From /home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/dataio/dataset_converter_lib.py:142: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.
2020-08-13 10:49:20,498 - tensorflow - WARNING - From /home/vpraveen/.cache/dazel/_dazel_vpraveen/715c8bafe7816f3bb6f309cd506049bb/execroot/ai_infra/bazel-out/k8-py3-fastbuild/bin/magnet/packages/iva/build_wheel.runfiles/ai_infra/iva/detectnet_v2/dataio/dataset_converter_lib.py:142: The name tf.python_io.TFRecordWriter is deprecated. Please use tf.io.TFRecordWriter instead.
/usr/local/lib/python3.6/dist-packages/iva/detectnet_v2/dataio/kitti_converter_lib.py:273: VisibleDeprecationWarning: Reading unicode strings without specifying the encoding argument is deprecated. Set the encoding, use None for the system default.
2020-08-13 10:49:20,511 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 1
2020-08-13 10:49:20,517 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 2
2020-08-13 10:49:20,523 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 3
2020-08-13 10:49:20,530 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 4
2020-08-13 10:49:20,537 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 5
2020-08-13 10:49:20,543 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 6
2020-08-13 10:49:20,549 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 7
2020-08-13 10:49:20,555 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 8
2020-08-13 10:49:20,561 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 0, shard 9
2020-08-13 10:49:20,572 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO -
Wrote the following numbers of objects:
b’with_mask’: 435
b’without_mask’: 66
b’mask_weared_incorrect’: 12
2020-08-13 10:49:20,572 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 0
2020-08-13 10:49:20,614 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 1
2020-08-13 10:49:20,656 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 2
2020-08-13 10:49:20,697 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 3
2020-08-13 10:49:20,740 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 4
2020-08-13 10:49:20,782 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 5
2020-08-13 10:49:20,823 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 6
2020-08-13 10:49:20,866 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 7
2020-08-13 10:49:20,906 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 8
2020-08-13 10:49:20,947 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Writing partition 1, shard 9
2020-08-13 10:49:20,988 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO -
Wrote the following numbers of objects:
b’with_mask’: 2266
b’without_mask’: 555
b’mask_weared_incorrect’: 90
2020-08-13 10:49:20,989 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Cumulative object statistics
2020-08-13 10:49:20,989 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO -
Wrote the following numbers of objects:
b’with_mask’: 2701
b’without_mask’: 621
b’mask_weared_incorrect’: 102
2020-08-13 10:49:20,989 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Class map.
Label in GT: Label in tfrecords file
b’with_mask’: b’with_mask’
b’without_mask’: b’without_mask’
b’mask_weared_incorrect’: b’mask_weared_incorrect’
For the dataset_config in the experiment_spec, please use labels in the tfrecords file, while writing the classmap.
2020-08-13 10:49:20,989 - iva.detectnet_v2.dataio.dataset_converter_lib - INFO - Tfrecords generation complete.
Data convert config :
kitti_config {
root_directory_path: “/workspace/tlt-experiments/data/data_with_eval/train”
image_dir_name: “images”
label_dir_name: “labels”
image_extension: “.jpg”
partition_mode: “random”
num_partitions: 2
val_split: 14
num_shards: 10
}
image_directory_path: “/workspace/tlt-experiments/data/data_with_eval/train”
Data convert command
tlt-dataset-convert -d converter_config/config_train.txt -o data/data_with_eval/train/tfrecords/
Detectnet v2 config :
random_seed: 42
dataset_config {
data_sources: {
tfrecords_path: “/workspace/tlt-experiments/data/data_with_eval/train/tfrecords*”
image_directory_path: “/workspace/tlt-experiments/data/data_with_eval/train/”
}
image_extension: “jpg”
target_class_mapping {
key: “with_mask”
value: “with_mask”
}
target_class_mapping {
key: “with_out_mask”
value: “with_out_mask”
}
target_class_mapping {
key: “mask_weared_incorrect”
value: “mask_weared_incorrect”
}
validation_fold: 0
}
model_config {
pretrained_model_file: “/workspace/tlt-experiments/tlt_peoplenet_vunpruned_v2.0/resnet34_peoplenet.tlt”
num_layers: 34
freeze_blocks: 0
arch: “resnet”
use_batch_norm: true
objective_set {
bbox {
scale: 35.0
offset: 0.5
}
cov {
}
}
training_precision {
backend_floatx: FLOAT32
}
}
training_config {
batch_size_per_gpu: 24
num_epochs: 120
learning_rate {
soft_start_annealing_schedule {
min_learning_rate: 5e-06
max_learning_rate: 0.0005
soft_start: 0.1
annealing: 0.7
}
}
regularizer {
type: L1
weight: 3e-09
}
optimizer {
adam {
epsilon: 9.9e-09
beta1: 0.9
beta2: 0.999
}
}
cost_scaling {
initial_exponent: 20.0
increment: 0.005
decrement: 1.0
}
checkpoint_interval: 10
}
augmentation_config {
preprocessing {
output_image_width: 960
output_image_height: 544
output_image_channel: 3
crop_right: 960
crop_bottom: 544
min_bbox_width: 1.0
min_bbox_height: 1.0
}
spatial_augmentation {
hflip_probability: 0.5
zoom_min: 1.0
zoom_max: 1.0
translate_max_x: 8.0
translate_max_y: 8.0
}
color_augmentation {
hue_rotation_max: 25.0
saturation_shift_max: 0.20000000298
contrast_scale_max: 0.10000000149
contrast_center: 0.5
}
}
postprocessing_config{
target_class_config{
key: “with_mask”
value: {
clustering_config {
coverage_threshold: 0.005
dbscan_eps: 0.265
dbscan_min_samples: 0.05
minimum_bounding_box_height: 4
}
}
}
target_class_config{
key: “with_out_mask”
value: {
clustering_config {
coverage_threshold: 0.005
dbscan_eps: 0.15
dbscan_min_samples: 0.05
minimum_bounding_box_height: 4
}
}
}
target_class_config{
key: “mask_weared_incorrect”
value: {
clustering_config {
coverage_threshold: 0.005
dbscan_eps: 0.15
dbscan_min_samples: 0.05
minimum_bounding_box_height: 2
}
}
}
}
TLT train command
tlt-train detectnet_v2 -e detect_net_config.txt -r output_dir/ -k tlt --gpus 1
Contents of tfrecords folder :
