I want to use tensorflow insinde a Jupyter notebook. However, running
import tensorflow as tf
in a Jupyter notebook immediately triggers a pop-up:
The kernel appears to have died. It will restart automatically.
This issue only began after updating to tensorflow 1.7. I had not used tensorflow for a few weeks so it might also be due to an update to anaconda 5.1 with Python 3.6.
I use a Mid-2010 MacBookPro with "High Sierra 10.13.4". Removing and reinstalling anaconda 5.1 with Python 3.6, followed by installation of tensorflow (and not a single other library) via
pip3 install --upgrade tensorflow
did not resolve the issue. I do not use an isolated environment. The "anaconda3" folder is not in my home folder but directly in "Macintosh HD".
Before reinstalling anaconda, I removed it via these instructions
https://docs.anaconda.com/anaconda/install/uninstall. I also did not try to run tensorflow outside Jupyter, simply because I do not know how. But even if I did, I would still like to use Jupyter.
I'm also running a Mid-2010 MacbookPro and have been facing the same issue. It seems the only solution is to downgrade to Tensorflow 1.5. You can do so by running the following:
pip3 uninstall tensorflow
pip3 install tensorflow==1.5
Credit given to the solution to this post.
I was facing the same issue with Tensorflow 2 '2.0.0-beta1'. I found out that when you have multiple notebooks with Tensorflow running, this problem occurs. Also, simply closing the unused notebook windows wont work, they're still running in the background, you'll have to 'Shutdown' the notebooks
Here are the steps to shutdown a notebooks:
> Go to Home (of Jupyter notebook)
> Select 'Running' tab
> Select the unused notebooks
> Click 'Shutdown' button
You will notice in the Jupyter Home that the active notebook icon is green while inactive ones are gray
I have also faced a similar issue.
I was using python 3.7 and Tensorflow version 1.5
For we moving to Python 3.5 solve the issue.
Related
I can not use Python in Jupyter (in VSCode) after updating my MacOS to Ventura 13.2 (22D49). It seems to be "import pymongo" that causes the error.
I tried following actions without success:
Use ipykernel but it didn't solve the problem.
according to this video
Uninstall VSCode
Uninstall Jupyter in VSCode
I have a MacBook Pro with M1 processor.
Happy to get some suggestions!
I don't know how much of a permanent solution this is, but what worked for me was going into VS Code Extensions, search for Jupyter (which should be installed already), then at the top I chose "Switch to Pre-release version". Then you have to reload the extension and it worked for me. I hope it works for you too!
I am having trouble running my import statement in VS code Jupyter. I split them into individual cells. I find when I run
import numpy as np
the cell hangs and I get a message
Connecting to kernel: Python 3.6.9: Waiting for Jupyter Session to be idle
How do I fix this?
To solve it, I uninstalled the extension Jupyter notebook (which requires a reload), and then installed it.
This may be related to the extended version. I hope this article is helpful to you.
Alright so this one surprised me..
I was using Jupyter-like code cells "#%%" (see docs) to run jupyter notebook in VSCode. And I ran into the same issue as OP.
The error disappeared when I renamed my file, from "inspect.py" to "tmp.py".
I found my solution was to select the correct version of Python. I had 2 choices for Python 3.6.9. I chose the one called "base(Python 3.6.9)" which had a different location to "Python 3.6.9" and the base version worked. Something odd if going on here, maybe I should remove the other version?
For me, upgrading ipython to version 7.34.0 (from 7.32.0) fixed it. I'm using jedi version 0.18.1. Related
Update: this broke again for me when I upgraded my virtual environment to Python 3.8. I just upgraded my ipykernel package and now the notebook runs.
This question already has answers here:
Fix not load dynamic library for Tensorflow GPU
(5 answers)
Closed 1 year ago.
I was originally running Tensorflow using PyCharm.
In PyCharm, the same phrase as the title did not appear.
But after I switched to VS Code and installed Python extension,
When I write and execute import tensorflow as tf, the error like the title appears repeatedly.
ImportError: Could not load dynamic library 'cudart64_110.dll'
Considering that there was no problem in PyCharm, it does not seem to be an environmental variable problem.
When I type the same command that was executed in VS Code in the command prompt window, another phrase appears,
"Connection failed because the target computer refused to connect."
My OS: Windows 10
I am using Anaconda, and I created a virtual environment.
vscode ver : 1.53.2
tensorflow ver : 2.4.1
CUDA : 11.2
cudnn : 8.1
It is due to tensorflow GPU support. Tensorflow now comes with GPU support and the system need graphics support and CUDA, CUDU installations. If you missed CUDA installation then you will get the above message. The latest version of tensorflow sometimes won't run without CUDA.
Try to install tensorflow 1.15 and python 3.7.4
https://www.python.org/ftp/python/3.7.4/python-3.7.4-amd64.exe
pip install tensorflow==1.15
NB: Normally tensorflow will run without cuda but the message will always shown in the prompt.
I would agree that this is due to your CUDA version, check the bottom of tensorflow GPU build config, it says for 2.4, you need CUDA 11.0 and cuDNN 8.0, which you have neither, in addition, you need MSVC 2019 to compile it.
Notice that for newer versions of tensorflow-gpu (>=2.3.0), conda will NOT download everything, you need to do them manually.
because it seems like all the evidence is pointing to GPU support problem, tensorflow-gpu might still run without using GPU, so it is possible that it was running on CPU when you use PyCharm,
I would suggest you double-check it runs as intended in PyCharm with
print(tf.config.list_physical_devices('GPU'))
or just simply reinstall everything
I copied "cudart64_110.dll" to the CUDA/v11.2/bin folder and it was resolved.
I have Python3 pre-installed on my personal laptop and now i have installed Jupyter Notebook using PIP but when i create a new notebook using python 3 kernel, it always gives orange highlighted message "Connecting to Kernel" but never connects.
OS - Windows 10, Personal laptop
Network - Wifi and not connecting through any proxy
Anti-virus - Uninstalled
Can someone help what could be the issue ?
I found the answer on stack overflow only in another thread.
Uninstall tornado 6
pip uninstall tornado
and reinstall tornado 5
pip install tornado==5.1.1
I had the same issue,i tried uninstalling tornado and reinstalling it ,i tried running it on different port , none of those worked.but i tried this and worked!!!!!
open anaconda ===> Home ====> find jupyter notebook ====> go to the gear icon (settings) ===> first remove it then you see there is a install button under it,after installation it worked for me!!!!
After having issues with this myself, and looking at the Tornado version, I found this wasn't the problem for me, instead after running the following in the relevant environment after installing ipykernel, this resolved the issue:
conda install -n base nb_conda_kernels
jupyter kernelspec list
Should list all kernels running on your system, make sure that the Python kernel path is accurate.
I would also try to run it over another port, in case if any network restrictions
I'm trying Anaconda/Spyder after using the standard python installation/IDLE for several years. The console (and variable explorer) becomes progressively slower during each work session, with every script and interactive command that I run. Eventually it takes minutes to get a response to simple commands like print, and then it stops responding completely and I shut it down. It's pretty disruptive to my work.
Mac OSX 10.13.6
conda version : 4.5.11
conda-build version : 3.15.1
python version : 3.7.0.final.0
qt 5.9.6
spyder 3.3.1
ipython 7.1.1
I do use matplotlib frequently. Whenever possible I use plt.ioff() at the beginning of a script and/or plt.close('all') at the end. Turning off matplotlib support completely in preferences doesn't help and it's not a permanent fix anyway. Python is not using anywhere near all my CPU or RAM.
As far as I've found out, the problem seems to be with the latest major release of the ipython kernel. Pinning the version of ipykernel to 4.10.0 solved it for me. To do so, add a line containing ipykernel 4.* to the file ~/anaconda3/conda-meta/pinned. Create the file if it does not exist yet. Then conda update ipykernel which will actually downgrade to 4.10.0. Worked for me!
you don't need to downgrade the entire spyder.
For me it worked with downgrading these two:
conda install ipython=6.4.0
conda install ipykernel=4.10.0
So this is not a totally satisfying solution but it seems to be a reasonable temporary fix. If you downgrade your spyder version to 3.1.X you will be able to use the python console rather than the Ipython console, and this seems to work for me. To do this just run:
conda install spyder=3.1
in the terminal while your environment of choice is active. This is not ideal, because the developers removed the python console for a specific reason as mentioned in this question, How to add python console in spyder.
Additionally downgrading this much may have unforeseen effects on other packages your code uses and may cause other issues but so far this is the only solution I was able to come up with.
#Sjoerd Thanks for your answer, this is what worked for me. To be still clearer, here's what I did in terminal on Anaconda 3 on High Sierra:
echo "ipykernel 4.10.0" >> ~/anaconda3/conda-meta/pinned
conda update anaconda
When I tried to update only ipykernel, I would have dependencies errors. By updating anaconda, everything is now up to date but ipykernel which is effectively at version 4.10.0. I didn't experience any hang since then.
#JJR4's answer didn't work for me (seems to work for others). Another workaround is to type %connect_info magic command into the IPython Spyder console which will display the JSON info to connect another console to the kernel.
The output of the magic command explains how to connect, but basically there are three options.
Paste the above JSON into a file, and connect with:
$> jupyter <app> --existing <file>
or, if you are local, you can connect with just:
$> jupyter <app> --existing <kernel>.json
or even just:
$> jupyter <app> --existing
if this is the most recent Jupyter kernel you have started.
I typically use qtconsole as the app. Once you enter your chosen connection method into the command line, a new console will open that is connected to the same kernel as the console in Spyder. For some reason, the new console does not lag even if the Spyder console does.
Enter everything in the new console, even runfile()
Hope this helps!
UPDATE:
Downgrading to ipykernel version 4.10 seems to solve the issue for me. Thanks to Carlos Cordoba - see GitHub conversation in link below.