J'essaye d'installer la version suivante de TensorFlow-GPU parce que l'auteur de gitrepo l'a suggéréici .
$ cat /usr/local/cuda/version.txt CUDA Version 10.0.130 $ pip list | grep tensorflow tensorflow-estimator 1.15.1 [jalal@goku examples]$ yum search wrapt Loaded plugins: aliases, changelog, copr, fastestmirror, kabi, langpacks, priorities, tmprepo, verify, versionlock Loading support for Red Hat kernel ABI Loading mirror speeds from cached hostfile * centos-sclo-rh: mirrors.seas.harvard.edu * centos-sclo-sclo: mirrors.seas.harvard.edu * remi-php70: mirror.pit.teraswitch.com * remi-php71: mirror.pit.teraswitch.com * remi-php73: mirror.pit.teraswitch.com * remi-safe: mirror.pit.teraswitch.com * ulyaoth: repos.ulyaoth.com * webtatic: us-east.repo.webtatic.com Atom 140/140 1075 packages excluded due to repository priority protections =============================================================================================== N/S matched: wrapt ================================================================================================ python-wrapt.x86_64 : A Python module for decorators, wrappers and monkey patching python36-wrapt.x86_64 : A Python module for decorators, wrappers and monkey patching Name and summary matches only, use "search all" for everything. [jalal@goku examples]$ sudo yum remove python-wrapt.x86_64 Loaded plugins: aliases, changelog, copr, fastestmirror, kabi, langpacks, priorities, tmprepo, verify, versionlock Loading support for Red Hat kernel ABI No Match for argument: python-wrapt.x86_64 No Packages marked for removal [jalal@goku examples]$ sudo yum remove python36-wrapt.x86_64 Loaded plugins: aliases, changelog, copr, fastestmirror, kabi, langpacks, priorities, tmprepo, verify, versionlock Loading support for Red Hat kernel ABI No Match for argument: python36-wrapt.x86_64 No Packages marked for removal
Comment résoudre ce problème d'installation?
Un peu d'informations sur mon système:
$ nvidia-smi Sat Dec 7 22:48:51 2019 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 418.67 Driver Version: 418.67 CUDA Version: 10.1 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | |===============================+======================+======================| | 0 GeForce GTX 108... Off | 00000000:05:00.0 On | N/A | | 0% 28C P0 61W / 250W | 1462MiB / 11178MiB | 0% Default | +-------------------------------+----------------------+----------------------+ | 1 GeForce GTX 108... Off | 00000000:06:00.0 Off | N/A | | 0% 28C P8 13W / 250W | 12MiB / 11178MiB | 0% Default | +-------------------------------+----------------------+----------------------+
J'ai les GPU suivants:
$ cat /proc/version
Linux version 3.10.0-1062.4.3.el7.x86_64 (mockbuild@kbuilder.bsys.centos.org) (gcc version 4.8.5 20150623 (Red Hat 4.8.5-39) (GCC) ) #1 SMP Wed Nov 13 23:58:53 UTC 2019
$ lsb_release -a
LSB Version: :core-4.1-amd64:core-4.1-noarch
Distributor ID: CentOS
Description: CentOS Linux release 7.7.1908 (Core)
Release: 7.7.1908
Codename: Core
$ python
Python 3.6.7 | packaged by conda-forge | (default, Feb 28 2019, 09:07:38)
[GCC 7.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ModuleNotFoundError: No module named 'tensorflow'
>>> import deepposekit
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages/deepposekit/__init__.py", line 20, in <module>
from deepposekit.io import TrainingGenerator, DataGenerator
File "/scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages/deepposekit/io/__init__.py", line 18, in <module>
from deepposekit.io.BaseGenerator import BaseGenerator
File "/scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages/deepposekit/io/BaseGenerator.py", line 16, in <module>
from tensorflow.keras.utils import Sequence
ModuleNotFoundError: No module named 'tensorflow'
Et voici ma version CUDA:
[jalal@goku examples]$ pip install tensorflow-gpu~=1.14 Collecting tensorflow-gpu~=1.14 Using cached https://files.pythonhosted.org/packages/a5/ad/933140e74973fb917a194ab814785e7c23680ca5dee6d663a509fe9579b6/tensorflow_gpu-1.15.0-cp36-cp36m-manylinux2010_x86_64.whl Processing /home/grad3/jalal/.cache/pip/wheels/d7/de/2e/efa132238792efb6459a96e85916ef8597fcb3d2ae51590dfd/wrapt-1.11.2-cp36-cp36m-linux_x86_64.whl Requirement already satisfied: protobuf>=3.6.1 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (3.11.0) Requirement already satisfied: tensorflow-estimator==1.15.1 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.15.1) Requirement already satisfied: six>=1.10.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.13.0) Requirement already satisfied: wheel>=0.26 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (0.33.6) Requirement already satisfied: tensorboard<1.16.0,>=1.15.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.15.0) Requirement already satisfied: termcolor>=1.1.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.1.0) Requirement already satisfied: opt-einsum>=2.3.2 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (3.1.0) Collecting keras-applications>=1.0.8 Using cached https://files.pythonhosted.org/packages/71/e3/19762fdfc62877ae9102edf6342d71b28fbfd9dea3d2f96a882ce099b03f/Keras_Applications-1.0.8-py3-none-any.whl Requirement already satisfied: keras-preprocessing>=1.0.5 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.0.5) Requirement already satisfied: grpcio>=1.8.6 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.25.0) Requirement already satisfied: absl-py>=0.7.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (0.8.1) Requirement already satisfied: google-pasta>=0.1.6 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (0.1.8) Processing /home/grad3/jalal/.cache/pip/wheels/5c/2e/7e/a1d4d4fcebe6c381f378ce7743a3ced3699feb89bcfbdadadd/gast-0.2.2-cp36-none-any.whl Requirement already satisfied: astor>=0.6.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (0.8.0) Requirement already satisfied: numpy<2.0,>=1.16.0 in /home/grad3/jalal/.local/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.17.4) Requirement already satisfied: setuptools in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from protobuf>=3.6.1->tensorflow-gpu~=1.14) (42.0.2) Requirement already satisfied: markdown>=2.6.8 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow-gpu~=1.14) (3.1.1) Requirement already satisfied: werkzeug>=0.11.15 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow-gpu~=1.14) (0.16.0) Requirement already satisfied: h5py in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from keras-applications>=1.0.8->tensorflow-gpu~=1.14) (2.7.1) Installing collected packages: wrapt, keras-applications, gast, tensorflow-gpu Found existing installation: wrapt 1.10.11 ERROR: Cannot uninstall 'wrapt'. It is a distutils installed project and thus we cannot accurately determine which files belong to it which would lead to only a partial uninstall.
7 Réponses :
Tout d'abord, procédez comme suit:
[jalal@goku examples]$ pip install tensorflow-gpu~=1.14
Collecting tensorflow-gpu~=1.14
Using cached https://files.pythonhosted.org/packages/a5/ad/933140e74973fb917a194ab814785e7c23680ca5dee6d663a509fe9579b6/tensorflow_gpu-1.15.0-cp36-cp36m-manylinux2010_x86_64.whl
Requirement already satisfied: numpy<2.0,>=1.16.0 in /home/grad3/jalal/.local/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.17.4)
Requirement already satisfied: google-pasta>=0.1.6 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (0.1.8)
Requirement already satisfied: wrapt>=1.11.1 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.11.2)
Requirement already satisfied: absl-py>=0.7.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (0.8.1)
Collecting keras-applications>=1.0.8
Using cached https://files.pythonhosted.org/packages/71/e3/19762fdfc62877ae9102edf6342d71b28fbfd9dea3d2f96a882ce099b03f/Keras_Applications-1.0.8-py3-none-any.whl
Requirement already satisfied: six>=1.10.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.13.0)
Requirement already satisfied: keras-preprocessing>=1.0.5 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.0.5)
Requirement already satisfied: opt-einsum>=2.3.2 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (3.1.0)
Requirement already satisfied: protobuf>=3.6.1 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (3.11.0)
Requirement already satisfied: tensorboard<1.16.0,>=1.15.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.15.0)
Requirement already satisfied: wheel>=0.26 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (0.33.6)
Requirement already satisfied: termcolor>=1.1.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.1.0)
Collecting tensorflow-estimator==1.15.1
Using cached https://files.pythonhosted.org/packages/de/62/2ee9cd74c9fa2fa450877847ba560b260f5d0fb70ee0595203082dafcc9d/tensorflow_estimator-1.15.1-py2.py3-none-any.whl
Processing /home/grad3/jalal/.cache/pip/wheels/5c/2e/7e/a1d4d4fcebe6c381f378ce7743a3ced3699feb89bcfbdadadd/gast-0.2.2-cp36-none-any.whl
Requirement already satisfied: astor>=0.6.0 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (0.8.0)
Requirement already satisfied: grpcio>=1.8.6 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorflow-gpu~=1.14) (1.25.0)
Requirement already satisfied: h5py in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from keras-applications>=1.0.8->tensorflow-gpu~=1.14) (2.7.1)
Requirement already satisfied: setuptools in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from protobuf>=3.6.1->tensorflow-gpu~=1.14) (42.0.2.post20191201)
Requirement already satisfied: werkzeug>=0.11.15 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow-gpu~=1.14) (0.16.0)
Requirement already satisfied: markdown>=2.6.8 in /scratch/sjn-p3/anaconda/anaconda3/lib/python3.6/site-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow-gpu~=1.14) (3.1.1)
Installing collected packages: keras-applications, tensorflow-estimator, gast, tensorflow-gpu
Found existing installation: Keras-Applications 1.0.6
Uninstalling Keras-Applications-1.0.6:
Successfully uninstalled Keras-Applications-1.0.6
Found existing installation: gast 0.3.2
Uninstalling gast-0.3.2:
Successfully uninstalled gast-0.3.2
Successfully installed gast-0.2.2 keras-applications-1.0.8 tensorflow-estimator-1.15.1 tensorflow-gpu-1.15.0
Puis,
$ conda update wrapt
Trouvez ce fichier .whl de cache dans votre /home/grad3/jalal/.cache/pip/wheels / ... Retirez-le.
Puis recommencez comme la procédure d'installation normale. Ça marche pour moi.
Mettre à jour Conda et essayer d'installer à nouveau Tensorflow a fonctionné pour moi:
conda update --all pip install tensorflow
Essayez d'utiliser les commandes suivantes:
pip install wrapt --upgrade --ignore-installed pip install tensorflow-gpu
Bonne chance.
pip install --upgrade pip conda update --all pip install tensorflow
Vous pouvez essayer les éléments suivants dans l'ordre:
pip install -U --ignore-installed wrapt pip install tensorflow
Supprimer le fichier wrapt-1.10.11.egg-info par la commande sudo rm /usr/lib/python3/dist-packages/wrapt-1.10.11.egg-info
Puis réinstallez tensorflow
vous devez
yum remove wrapt, car il n'a pas été installé avec pip, il ne peut pas être mis à niveau avec pip.Merci William mais j'ai eu cette erreur:
$ sudo yum remove wrapt [sudo] password for jalal: Loaded plugins: aliases, changelog, copr, fastestmirror, kabi, langpacks, priorities, tmprepo, verify, versionlock Loading support for Red Hat kernel ABI No Match for argument: wrapt No Packages marked for removalJe ne suis pas sur mon ordinateur de travail, je ne peux donc pas vérifier le nom correct du package. Il pourrait être
yum remove python-wrapt. (Et après l'avoir supprimé, vous devrez réessayer le pip install tensorflow-gpu)veuillez vérifier la question mise à jour. J'ai recherché le wrapt mais je n'ai pas pu le trouver installé.
il est dans la liste pip mais dois-je pip le désinstaller?
$ pip list | grep wrapt wrapt 1.10.11ok, vous utilisez conda. Je n'ai pas remarqué cela.
conda update wraptouconda update --allPour plus d'informations: github.com/tensorflow/tensorflow/issues/30191$ pip uninstall wrapt ERREUR: Impossible de désinstaller «wrapt». Il s'agit d'un projet installé distutils et nous ne pouvons donc pas déterminer avec précision quels fichiers lui appartiennent, ce qui ne conduirait qu'à une désinstallation partielle.
@ WilliamD.Irons c'est drôle mais ma
conda update --allest bloqué depuis 30 minutes ici: unix.stackexchange.com/questions/556133 / ...Je ne suis pas sûr de ça. pip ne désinstallera pas quelque chose qu'il n'a pas installé. C'est pourquoi vous devez utiliser conda pour le mettre à jour ou le supprimer si c'est ainsi que wrapt a été installé.