# I run tensorflow ,by python3.but CUDA runtime implicit initialization on GPU:0 failed internal **URL:** \ **Category:** Jetson TX2\ **Created:** [September 19, 2018, 2:42pm UTC](https://forums.developer.nvidia.com/t/i-run-tensorflow-by-python3-but-cuda-runtime-implicit-initialization-on-gpu-0-failed-internal/65290 "2018-09-19T14:42:37Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![njlw](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@njlw](https://forums.developer.nvidia.com/u/njlw)\ **Post date:** [September 19, 2018, 2:42pm UTC](https://forums.developer.nvidia.com/t/i-run-tensorflow-by-python3-but-cuda-runtime-implicit-initialization-on-gpu-0-failed-internal/65290/1 "2018-09-19T14:42:37Z")
I run tensorflow ,by python3.but:CUDA runtime implicit initialization on GPU:0 failed internal the log is this: Python 3.5.2 (default, Nov 23 2017, 16:37:01) [GCC 5.4.0 20160609] on linux Type “help”, “copyright”, “credits” or “license” for more information. > > > import tensorflow as tf > > > > > > sess = tf.Session() > > > 2018-09-19 14:39:28.372649: I tensorflow/stream\_executor/cuda/cuda\_gpu\_executor.cc:864] ARM64 does not support NUMA - returning NUMA node zero > > > 2018-09-19 14:39:28.372904: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:1392] Found device 0 with properties: > > > name: NVIDIA Tegra X2 major: 6 minor: 2 memoryClockRate(GHz): 1.3005 > > > pciBusID: 0000:00:00.0 > > > totalMemory: 7.66GiB freeMemory: 4.73GiB > > > 2018-09-19 14:39:28.373072: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:1471] Adding visible gpu devices: 0 2018-09-19 14:39:30.970862: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:952] Device interconnect StreamExecutor with strength 1 edge matrix: 2018-09-19 14:39:30.970954: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:958] 0 2018-09-19 14:39:30.971010: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:971] 0: N 2018-09-19 14:39:30.971201: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:1084] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 4339 MB memory) → physical GPU (device: 0, name: NVIDIA Tegra X2, pci bus id: 0000:00:00.0, compute capability: 6.2) > > > hello = tf.constant(‘Hello, TensorFlow!’) > > > sess = tf.Session() > > > 2018-09-19 14:39:41.348955: I tensorflow/core/common\_runtime/gpu/gpu\_device.cc:1471] Adding visible gpu devices: 0 > > > 2018-09-19 14:39:41.349823: E tensorflow/core/common\_runtime/direct\_session.cc:158] Internal: CUDA runtime implicit initialization on GPU:0 failed. Status: unknown error > > > Traceback (most recent call last): > > > File “”, line 1, in > > > File “/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py”, line 1563, in **init** > > > super(Session, self). **init** (target, graph, config=config) > > > File “/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py”, line 633, in **init** > > > self.\_session = tf\_session.TF\_NewSession(self.\_graph.\_c\_graph, opts) > > > tensorflow.python.framework.errors\_impl.InternalError: Failed to create session. ---
**Author:** ![AastaLLL](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aastalll/32/14043_2.png) [@AastaLLL](https://forums.developer.nvidia.com/u/AastaLLL)\ **Post date:** [September 20, 2018, 5:54am UTC](https://forums.developer.nvidia.com/t/i-run-tensorflow-by-python3-but-cuda-runtime-implicit-initialization-on-gpu-0-failed-internal/65290/2 "2018-09-20T05:54:49Z")
Hi, Please noticed that there are some dependencies on GPU architecture and python version. It’s recommended to install our TensorFlow package for python3. [url][https://devtalk.nvidia.com/default/topic/1038957/jetson-tx2/tensorflow-for-jetson-tx2-/[/url]](https://devtalk.nvidia.com/default/topic/1038957/jetson-tx2/tensorflow-for-jetson-tx2-/%5B/url%5D) Please let us know if you still meet an error with our official package. Thanks. ---
**Author:** ![njlw](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@njlw](https://forums.developer.nvidia.com/u/njlw)\ **Post date:** [September 20, 2018, 7:26am UTC](https://forums.developer.nvidia.com/t/i-run-tensorflow-by-python3-but-cuda-runtime-implicit-initialization-on-gpu-0-failed-internal/65290/3 "2018-09-20T07:26:03Z")
Did you mean: reinstall again? ---
**Author:** ![AastaLLL](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aastalll/32/14043_2.png) [@AastaLLL](https://forums.developer.nvidia.com/u/AastaLLL)\ **Post date:** [September 26, 2018, 5:24am UTC](https://forums.developer.nvidia.com/t/i-run-tensorflow-by-python3-but-cuda-runtime-implicit-initialization-on-gpu-0-failed-internal/65290/4 "2018-09-26T05:24:02Z")
YES. It’s recommended to try our official TensorFlow package. Thanks. ---
**Author:** ![nadeemm](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/nadeemm/32/367020_2.png) [@nadeemm](https://forums.developer.nvidia.com/u/nadeemm)\ **Post date:** [October 18, 2021, 6:30pm UTC](https://forums.developer.nvidia.com/t/i-run-tensorflow-by-python3-but-cuda-runtime-implicit-initialization-on-gpu-0-failed-internal/65290/5 "2021-10-18T18:30:22Z")
Closed on October 18, 2021, 6:30pm UTC