I am setting setting up TensorFlow Object Detection API for retraining of a pre-trained model on Jetson Orin Development Kit.
Following are some of the details:
- OS: Linux ubuntu 5.10.104-tegra #1 SMP PREEMPT Tue Jan 24 15:09:44 PST 2023 aarch64 aarch64 aarch64 GNU/Linux
- JetPack 5.1
- Python 3.8.10
- tensorflow version: 2.11.0
- tesorflow gpu support is True and can be verified by the following output:
f.test.is_gpu_available()
WARNING:tensorflow:From <stdin>:1: is_gpu_available (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.config.list_physical_devices('GPU')` instead.
2023-03-24 12:29:52.194575: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:984] could not open file to read NUMA node: /sys/bus/pci/devices/0000:00:00.0/numa_node
Your kernel may have been built without NUMA support.
2023-03-24 12:29:52.239510: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:984] could not open file to read NUMA node: /sys/bus/pci/devices/0000:00:00.0/numa_node
Your kernel may have been built without NUMA support.
2023-03-24 12:29:52.239714: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:984] could not open file to read NUMA node: /sys/bus/pci/devices/0000:00:00.0/numa_node
Your kernel may have been built without NUMA support.
2023-03-24 12:29:52.865652: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:984] could not open file to read NUMA node: /sys/bus/pci/devices/0000:00:00.0/numa_node
Your kernel may have been built without NUMA support.
2023-03-24 12:29:52.865972: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:984] could not open file to read NUMA node: /sys/bus/pci/devices/0000:00:00.0/numa_node
Your kernel may have been built without NUMA support.
2023-03-24 12:29:52.866044: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Could not identify NUMA node of platform GPU id 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2023-03-24 12:29:52.866163: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:984] could not open file to read NUMA node: /sys/bus/pci/devices/0000:00:00.0/numa_node
Your kernel may have been built without NUMA support.
2023-03-24 12:29:52.866298: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1616] Created device /device:GPU:0 with 19313 MB memory: -> device: 0, name: Orin, pci bus id: 0000:00:00.0, compute capability: 8.7
True
When I try to test that the Object Detection API installed using the link [Installation — TensorFlow 2 Object Detection API tutorial documentation](https://Object Detection API Installation).
Running the python object_detection/builders/model_builder_tf2_test.py results into following error:
/home/ki/.local/lib/python3.8/site-packages/requests/__init__.py:102: RequestsDependencyWarning: urllib3 (1.26.15) or chardet (5.1.0)/charset_normalizer (2.0.12) doesn't match a supported version!
warnings.warn("urllib3 ({}) or chardet ({})/charset_normalizer ({}) doesn't match a supported "
Traceback (most recent call last):
File "builders/model_builder_tf2_test.py", line 24, in <module>
from object_detection.builders import model_builder
File "/home/ki/.local/lib/python3.8/site-packages/object_detection/builders/model_builder.py", line 22, in <module>
from object_detection.builders import box_predictor_builder
File "/home/ki/.local/lib/python3.8/site-packages/object_detection/builders/box_predictor_builder.py", line 20, in <module>
from object_detection.predictors import convolutional_box_predictor
File "/home/ki/.local/lib/python3.8/site-packages/object_detection/predictors/convolutional_box_predictor.py", line 22, in <module>
slim = tf.contrib.slim
AttributeError: module 'tensorflow' has no attribute 'contrib'
Is there someone else encountering the same problem or can suggest any solution?
Thanks in advance,
Saqib