When I try to download tensorflow,it always happen errors,sush as [u]requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://developer.download.nvidia.cn/compute/redist/jp/v42/termcolor/[/u]
how can I slove it?
indeed this fixed the issue :
sudo pip3 install --pre --no-cache-dir --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v42 tensorflow-gpu
How can I install tensorflow for python 2.7 on jetson nano?
How can I install tensorflow for python 2.7 on jetson nano?
I’m already using Python3.7.4
Any build for Python3.7?
Cheers
Pei
You need to build from source to use with python 2. I am not sure if atm, Nvidia provides an official wheel for python2.
i have built my TF 1.12 from source for python 2
I followed the above instructions to install Tensorflow in a virtual environment created with Anaconda. The installation was successful but when I ran “import tensorflow as tf” , I got an error message “ModuleNotFoundError: No module named ‘tensorflow’”
Is there something am doing wrong please?
installed it
dlinano@jetson-nano:~$
dlinano@jetson-nano:~$ sudo -H pip3 install -U pip
Retrying (Retry(total=4, connect=None, read=None, redirect=None, status=None)) after connection broken by ‘NewConnectionError(’<urllib3.connection.VerifiedHTTPSConnection object at 0x7f8ce44710>: Failed to establish a new connection: [Errno -2] Name or service not known’,)': /simple/pip/
Retrying (Retry(total=3, connect=None, read=None, redirect=None, status=None)) after connection broken by ‘NewConnectionError(’<urllib3.connection.VerifiedHTTPSConnection object at 0x7f8ce44828>: Failed to establish a new connection:
I had problem with “sudo -H pip3 install -U pip”?
When I verify the installation as in https://docs.nvidia.com/deeplearning/frameworks/install-tf-jetson-platform/index.html#verify , I get numpy FutureWarnings:
dlinano@jetson-nano:~$ python3
Python 3.6.8 (default, Jan 14 2019, 11:02:34)
[GCC 8.0.1 20180414 (experimental) [trunk revision 259383]] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow
2019-08-21 08:02:59.283722: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudart.so.10.0
/home/dlinano/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
np_resource = np.dtype([("resource", np.ubyte, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
_np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/dlinano/.local/lib/python3.6/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
np_resource = np.dtype([("resource", np.ubyte, 1)])
>>>
Is this considered as an error? And how to get rid of it?
This seems to work:
dlinano@jetson-nano:~$ python3 -m pip show tensorflow-gpu
Name: tensorflow-gpu
Version: 1.14.0+nv19.7
Summary: TensorFlow is an open source machine learning framework for everyone.
Home-page: https://www.tensorflow.org/
Author: Google Inc.
Author-email: packages@tensorflow.org
License: Apache 2.0
Location: /home/dlinano/.local/lib/python3.6/site-packages
Requires: tensorboard, keras-applications, absl-py, gast, keras-preprocessing, tensorflow-estimator, wheel, google-pasta, numpy, six, termcolor, protobuf, astor, wrapt, grpcio
Required-by:
Tensorflow 2.0 RC 1 (tensorflow==2.0.0-rc1) was released 36 hours ago. Can I check if anyone is attempting to build from source for the Jetson Nano ? Or is NVIDIA releasing an updated version for an easy install for all of us ? =) THANKS
I got an error message running
pip3 install --pre --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v42 tensorflow-gpu==1.14.0+nv19.7
Collecting tensorflow-gpu==1.14.0+nv19.7
Downloading https://developer.download.nvidia.com/compute/redist/jp/v42/tensorflow-gpu/tensorflow_gpu-1.14.0+nv19.7-cp36-cp36m-linux_aarch64.whl (225.3MB)
100% |████████████████████████████████| 225.3MB 2.2kB/s
Requirement already satisfied: gast>=0.2.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.14.0+nv19.7)
Requirement already satisfied: wrapt>=1.11.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.14.0+nv19.7)
Requirement already satisfied: astor>=0.6.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-gpu==1.14.0+nv19.7)
Collecting tensorflow-estimator<1.15.0rc0,>=1.14.0rc0 (from tensorflow-gpu==1.14.0+nv19.7)
Exception:
...
requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://developer.download.nvidia.com/compute/redist/jp/v42/tensorflow-estimator/
But finally I successfully installed tensorflow-gpu by running the following commands:
sudo -H pip3 install tensorflow-estimator==1.14.0
sudo -H pip3 install tensorboard==1.14.0
and then
sudo -H pip3 install --pre --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v42 tensorflow-gpu==1.14.0+nv19.7
SInce tensorflow-gpu depends on tensorflow-estimator and tensorboard, while unfortunately these two packages are
not available under url https://developer.download.nvidia.com/compute/redist/jp/v42
When is the official release planned for Jetson compatible Tensorflow 2.0 from Nvidia?
I tried to build tensorflow-2.0.0 from source and install it on Jetson Nano. It appeared to work. Please refer to my blog post for details. (As before, I’ve put all necessary steps into a script so it should be very easy to follow.)
“Building TensorFlow 2.0.0 on Jetson Nano”: [url]https://jkjung-avt.github.io/build-tensorflow-2.0.0/[/url]
When will Tensorflow 2.x and Keras be available via the SDK Manager for the Jetson Nano?
This would centralize and simplify a lot of configuration error related questions.
Thanks
Kent Smith
I wish NVIDIA would just post a dockerhub image of a good-to-go tensorflow environment with all the bells and whistles and dependent libraries taken care of, and we just run it under docker. I do this on my workstations with other NVIDIA cards, seems like docker already works on Jetson Nano so what’s the barrier? Makes life a million times easier.
I have successfully build Tensorflow 2.0 from source on NANO. Here’s how I did the build. It requires alot of patience. Enjoy. =)
Have you tested it in order to check perfomance?
Hi,
I am new to this field. I am a bit confused about TensorFlow and TensorFlow Lite. I have installed the Tensorflow GPUs version in my nano. I have some very basic questions.
- For my object detection program what should I use TF or TF Lite in Jetson Nano?
- Is installing TF gpu version same as installing TF in gpu ? If yes, how do I install TF Lite?
- I have a tensor flow model. Should I convert it to TF Lite model in my windows laptop (Dell g15) or get it converted in nano?
Thanks in advance.
I have encountered a similar issue as d.a.mahabiersing, Is this considered as an error? And how to get rid of it?
Python 3.6.8 (default, Oct 7 2019, 12:59:55)
[GCC 8.3.0] on linux
Type “help”, “copyright”, “credits” or “license” for more information.
import tensorflow as tf
2019-10-16 00:22:26.240200: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudart.so.10.0
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_qint8 = np.dtype([(“qint8”, np.int8, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_quint8 = np.dtype([(“quint8”, np.uint8, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_qint16 = np.dtype([(“qint16”, np.int16, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_quint16 = np.dtype([(“quint16”, np.uint16, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_qint32 = np.dtype([(“qint32”, np.int32, 1)])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
np_resource = np.dtype([(“resource”, np.ubyte, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_qint8 = np.dtype([(“qint8”, np.int8, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_quint8 = np.dtype([(“quint8”, np.uint8, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_qint16 = np.dtype([(“qint16”, np.int16, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_quint16 = np.dtype([(“quint16”, np.uint16, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
_np_qint32 = np.dtype([(“qint32”, np.int32, 1)])
/usr/local/lib/python3.6/dist-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or ‘1type’ as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / ‘(1,)type’.
np_resource = np.dtype([(“resource”, np.ubyte, 1)])
$ python3 -m pip show tensorflow-gpu
Name: tensorflow-gpu
Version: 1.14.0+nv19.7
Summary: TensorFlow is an open source machine learning framework for everyone.
Home-page: https://www.tensorflow.org/
Author: Google Inc.
Author-email: packages@tensorflow.org
License: Apache 2.0
Location: /usr/local/lib/python3.6/dist-packages
Requires: numpy, keras-applications, wrapt, grpcio, tensorboard, termcolor, keras-preprocessing, six, protobuf, gast, google-pasta, absl-py, astor, wheel, tensorflow-estimator
Required-by: