A number of problems show up when attempting to install Tensorflow-gpu.
(First Problem): solving environment on anaconda fail (both initial frozen solve and repodata from current repodata).
(Second Problem main one): both CUDA and cuDNN are installed (usually they are a part of the installation of Tensorflow-gpu since v 2.2). This is the main problem, since although the installation using conda seems to success in fact when try to work with the tensorflow and import it later, a group of errors shows up of missing libraries. As seen with the screenshot, no CUDA or cuDNN available in the packagae plan for installation.
Attempt to fix it by reinstalling Anaconda didn't work out, also I reinstall Windows10 without success. Highlighting, it used to work and installed by this method normally before.
please advise ....
open anaconda navigator, at the top right there might be an update button. Update anaconda and then try install from the navigator. or you can try:
conda install -c anaconda tensorflow-gpu
Related
first time poster
I am using the Climada modelling tool with Spyder as my IDE of choice through Anaconda. I have created a new environment as per the Climada installation instructions and need to open Spyder to complete the final stages of the installation by running tests. Anyway...onto the problem.
I uninstalled Anaconda earlier, and then did a reinstall. Unfortunately after the reinstall Spyder was not opening in the Navigator, so I tried it through the prompt instead, and received the following message:
Terminal message (apologies, reputation is not high enough to display image)
I think it could be duplicate files leftover in my "AppData\Roaming" folders from my earlier uninstallation of Anaconda that is causing a few issues, but I'm not 100% sure.
Would anyone be able to confirm the problem/provide some help as to what I need to do to overcome this issue? At the moment I'm reluctant to delete any files that may need to be kept
Thanks in advance!
The conda environment for CLIMADA is known to create some issues with Spyder, although I am not sure what the exact problem with your particular issue is.
I can recommend a few things to try (also explained in the docs here):
First step, deinstall spyder in the environment and then:
Install spyder in the conda environment using pip instead of conda
conda activate climada_env
pip install spyder
Install spyder 4 instead of 5
Install spyder in a separate conda environment and configure it to use the climada_env (similarly as recommend by Daniel Althviz).
I'm using Anaconda 3.8.5 on Windows 7.
The problem: I was trying to install a Google Colab module locally, because I wanted to run this notebook on a local runtime, and it needed a module called "google-colab" (evidently this is a lost cause, as the module is not open source). I attempted to install it using conda install -c conda-forge google-colab, which I found here.
It appeared to install something successfully, although the notebook on colab still didn't work. After installing, Anaconda Navigator refused to launch. A command prompt popped up and disappeared instantly, and nothing else happened. I uninstalled, I reinstalled. I followed this procedure. Nothing changed at all.
When I installed Anaconda on a virtual machine with a fresh install of Windows 7 and attempted that same janky install, the exact same thing happened. Any idea how I can get Anaconda navigator back? Anaconda command prompt still works. Thank you!
I just installed anaconda ver5.3 which uses python v3.7
I ran the command;
conda update anaconda
The command hangs there forever. There is no error message. No hint what is wrong. Has anyone encountered similar problems? How did you solve it?
I am using Windows 10 and was previously using anaconda ver5.2 which uses python v3.6.
Try running
conda update anaconda --verbose
to get some insight. Adding --verbose multiple times increases logging to INFO, DEBUG and then TRACE level.
https://conda.io/docs/commands/conda-update.html
I think this could solve your problem to update anaconda, and yes it seems like a fairly common issue:
1) install the latest miniconda3 from here
2) make sure you run anaconda prompt as admin
3) update to latest version of conda
conda update conda
4) install anaconda
conda install anaconda
5) install navigator
conda install anaconda-navigator
6) verify conda is installed and check package number
conda info
View packages you installed in anaconda docs page here
I encountered exactly the same problem on Windows 10.
This problem has been fixed with the latest anaconda release ver5.3.1. You are using ver5.3. Please use the latest version. It should solve your problem.
I have Anaconda 4.4.0 (Windows, Python 3.6., 64 bit).
I would like to upgrade to latest Anaconda 5.0.1
Few options:
Download the full installer and run it
From existing installation (of 4.4.0) run "conda update --all"
From existing installation run "conda update anaconda"
What is the tradeoff among these options? What is the recommended one?
Download the full installer: Provided that you uninstall your existing Anaconda, this method will be least likely to cause upgrade problems. It will also probably be slower. Note that I think you should uninstall the old Anaconda so that you don't end up with two conda[.exe] files, two Anaconda Prompt shortcuts, and so forth. You may end up trying to install a package with the wrong conda and be very confused about what's happening.
conda update --all: This will update all of your packages in the environment to their latest version, regardless of their version in the Anaconda installer. This is not recommended because you will end up with package versions that are different from the ones in the Anaconda installer and you may end up with an error message about packages that are incompatible.
conda update anaconda: This will update the "metapackage" called anaconda to the latest version. This package has dependencies on specific versions of all of the packages and Anaconda (the company) give some assurance that these will all work together. So, updating the anaconda package will update all your packages to the version used in the latest version of the Anaconda installer.
My suggestion (based on some experience, I am not an employee of Anaconda) would be to try #3 and if it fails, try #1.
I was trying to install h2o (3.10.0.9) on Anaconda 3 in windows 10, 64 bit. The installation seemed to proceeded, and after it shows
"executing transaction:done"
what i got in the prompt window were a few command lines:
(base) C:\WINDOWS\system32>set "JAVA_HOME="
(base) C:\WINDOWS\system32>set "JAVA_HOME_CONDA_BACKUP="
(base) C:\WINDOWS\system32>set "JAVA_HOME=C:\Users\Xu\Anaconda3\Library"
And at this point, I still cannot import h2o in jupyter notebook (no module named h2o). Note that the installation of h2o and installation of openjdk was done at the same time. I downloaded JAVA outside anaconda, and did some manual changes in the environmental variables, but it seems that this is not the problem.
Anaconda Navigator shows that h2o is installed, while h2o-py is not. Now, the problem with h2o-py is that if I install it, it will show this error:
UnsatisfiableError: The following specifications were found to be in conflict:
- h2o-py
- zict
Use "conda info <package>" to see the dependencies for each package.
Now, I checked all the dependencies:
colorama
future
patsy
python 3.5*
requests
tabulate
I installed future, and then tried installing tabulate, but it also shows the same UnsatisfiableError ... in conflict with zict. Then I removed zict package, and then both the installations of tabulate and h2o-py show another similar error, but this time in conflict with xlrd package. It took me too long time that I really don't want to waste any more on this.
Could anybody tell to me what I can do in order to be able to simply use h2o in anaconda? I would very much appreciate it!
Though it's too late, may be helpful for someone else.
I had the same problem and solved it by installing 'h2o' using pip command:
pip install h2o
or if you are in IPython environment:
!pip install h2o
I had a similar issue on macOS, tried to install using conda install -c anaconda h2o as advised in anaconda documentation However, it did not reflect in my JupyterLab.
I finally resolved the issue by using !pip install h2o