I have Anaconda and Visual Studio Code installed on my computer. My default terminal for VS Code is Git Bash. When I open a new terminal in VSCode, it immediately runs the following commands:
C:/Users/ethan/AppData/Local/Continuum/anaconda3/Scripts/activate
conda activate base
The second of these commands gives the following error:
CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
If using 'conda activate' from a batch script, change your
invocation to 'CALL conda.bat activate'.
I have tried running conda init bash and conda init --all both inside the VSCode terminal, and inside Git Bash. It handles that command fine, but it doesn't solve my problem. I don't know if the second line of the error applies to me, but even if it did, I don't know how to change the command being called because it is done automatically by VSCode. This error occurs every time I launch a terminal in VSCode (even if I don't have any python files present in my workspace), and it happens both when I launch VSCode from the launch button in Anaconda Navigator and when I launch VSCode by itself.
I had the same issue. For me, easily resolved by launching VSC from the conda window.
Specifically, open your cmd prompt (for me, Anaconda Prompt), activate the environment using 'conda activate [envname]'. Then just run the command 'code'. This will launch VS Code with the activated environment and associated variables. From there, the debug works as expected.
I had the same issue, I've fixed it by adding the Python.CondaPath in settings.
Press Ctrl + Shift + P and select Terminal Configuration. Search for python.conda, and paste your conda path for example. C:\ProgramData\Anaconda3\Scripts\conda.exe
This will fix your issue.
In VS code settings, search for "terminal.integrated.shellArgs.windows", then click "Edit in settings.json". For me, this opened "%APPDATA%\Code\User\settings.json".
I set "terminal.integrated.shellArgs.windows": "-i -l" and this fixed it for me. My file:
{
"terminal.integrated.shell.windows": "C:\\Program Files\\Git\\bin\\bash.exe",
"terminal.integrated.shellArgs.windows": "-i -l"
}
Since conda activate command would cause CommandNotFoundError, use source activate command instead. It works the same.
I had the same issue. I got the following error:
CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
If using 'conda activate' from a batch script, change your invocation to 'CALL conda.bat activate'.
To initialize your shell, run
$ conda init <SHELL_NAME>
Currently supported shells are:
- bash
- cmd.exe
- fish
- tcsh
- xonsh
- zsh
- powershell
See 'conda init --help' for more information and options.
IMPORTANT: You may need to close and restart your shell after running 'conda init'.
I solved it by manully run
source activate base
and Ta-da~ DONE!
This trick is simple, while needed every time you open a new git-bash terminal in VSCode.
This worked for me go to setting and search for
terminal.integrated.shellargs.windows
then choose Command Prompt like this
I had exactly the same error as you. I solved it with a tip from a Python course in Udacity
Open Git Bash command line (NOT within VSC terminal) and go to your home folder, e.g., /c/Users/arman. Then run the following two commands but replace [YOUR_PATH] with your Anaconda installation folder
echo 'export PATH="$PATH:[YOUR_PATH]:[YOUR_PATH]/Scripts"' >> .bashrc
echo 'alias python="winpty python.exe"' >> .bashrc
For example in my case, as I have miniconda, I executed:
echo 'export PATH="$PATH:/c/Users/arman/Miniconda3:/c/Users/arman/Miniconda3/Scripts"' >> .bashrc
echo 'alias python="winpty python.exe"' >> .bashrc
After this executing those lines, i.e., creating the .bashrc file, then run:
source .bashrc
Afterwards, open VSC and try running or debugging a python program. It worked for me!
I solved this issue by using Powershell. Start the Powershell as Administrator and then type
set-ExecutionPolicy RemoteSigned
Say yes if it asks a confirmation. Now, VSCode debugger option can be used with Python.
When I changed my default terminal in VS Code to cmd.exe I got conda to work properly for me.
None of these worked for me. In the end I changed my default VScode terminal to cmd instead of Powershell in the terminal default settings which was None for Windows. I had already added conda.exe to my path and changed the permissions as described above.
For some reason the activate.bat file was now found in the path which activated my virtual env.
This answer is dedicated to Windows 10/11 users, based on PowerShell VS Code integrated terminal, and assumes using miniconda but the same holds for conda under the previous environment.
From the start menu type or search for Anaconda Powershell Prompt then right-click to open its file location. You should see the PowerShell shortcut. Right-click and open properties. In the target bar try to find C:\Users\username\miniconda3\shell\condabin\conda-hook.ps1' ; PowerShell script (.ps1). Using file explorer locate C:\Users\username\miniconda3\shell\condabin. Copy the content of that script.
In the default Documents folder create a folder named WindowsPowerShell if not exists and edit or create a new Microsoft.VSCode_profile.ps1 script in it. Lastly, paste by appending the content of the former script.
You could do the same for the external PowerShell but the script under the path C:\Users\username\Documents\WindowsPowerShell should be named Microsoft.PowerShell_profile.ps1.
In other words, the PowerShell profile script enables the creation of aliases and defining functions that load every time you launch the shell. However, the changes take effect after restarting VS Code and/or PowerShell.
I installed python3 in my macbook.
but when I open terminal, type which python, it returns
/Library/Frameworks/Python.framework/Versions/2.7/bin/python
So how can I change to the python3 for temporal use? I don't want to change the system default.
The reason why I change this is because I installed jupyter as pip3 and I want to open ipython file, when I type jupyter notebook, it says command not found, so I realise it because the python versions.
To change to python3, you can use the following command in terminal
alias python='python3'
you can also add this into ~/.bash_profile file
Edit you ~/.bash_profile and add the following line:
export PATH="$PATH:/path/to/your/Python/3.6/bin"
after that execute this once
source ~/.bash_profile
or restart the terminal
Does bash report that the command is not found when you enter 'jupyter notebook' as a command?
Pip should automatically make Jupyter available in $PATH but this may not be the case here if 'jupyter' is not a recognised command
Conversely if #Kallz's answer was successful then:
If you were looking for a consistent temporal environment for Jupyter where Jupyter runs in version x and you terminal shell/all other directories run version y, then you should refer to this answer about pyenv, which helps manage python versions and create Python 'environments'
If you can determine the specific directory location which executes Jupyter (I have no specific knowledge of Jupyter myself to help with this), then you can set a 'local' pyenv for the executing file location and a 'global' env for the version you want to use for shell etc., thus constraining to the correct version for both purposes.
See a setup tutorial here
I installed Anaconda and can run Python, so I assume that I installed it correctly. Following this introductory documentation, I am trying to install Python v3.3, so I am copying and pasting the following line into my console:
conda create -n py33 python=3.3 anaconda
However, that gives me an error:
-bash: conda: command not found
What do I need to do to run Conda?
I am working on a Linux system.
you might want to try this:
for anaconda 2 :
export PATH=~/anaconda2/bin:$PATH
for anaconda 3 :
export PATH=~/anaconda3/bin:$PATH
for anaconda 4 :
Use the Anaconda Prompt
and then
conda --version
to confirm that it worked.
The export PATH=~/anaconda3/bin:$PATH works but stops when you exit the terminal in order change that you have to run sudo nano ~/.bashrc and then copy the path into the file and save it after that you activate the changes using source .bashrc.
check with conda install anaconda-navigator
if not installed follow the anaconda install instructions again
follow along with this video
https://youtu.be/Pr25JlaXhpc
It turns out that I had not set the path.
To do so, I first had to edit .bash_profile (I downloaded it to my local desktop to do that, I do not know how to text edit a file from linux)
Then add this to .bash_profile:
PATH=$PATH:$HOME/anaconda/bin
Use conda init
As pointed out in a different answer, manually adding Conda on $PATH is no longer recommended as of v4.4.0 (see Release Notes). Furthermore, since Conda v4.6 new functionality to manage shell initialization via the conda init command was introduced. Hence, the updated recommendation is to run
Linux/UNIX (OS X < 10.15)
./anaconda3/bin/conda init
Mac OS X >= 10.15
./anaconda3/bin/conda init zsh
Windows
./anaconda3/Scripts/conda.exe init
You must launch a new shell or source your init file (e.g., source .bashrc) for the changes to take effect.
Details. For an in-depth look at exactly what conda init does see this answer.
Alternative shells
You may need to explicitly identify your shell to Conda. For example, if you run zsh (Mac OS X 10.15+ default) instead of bash then you would run
./anaconda3/bin/conda init zsh
Please see ./anaconda3/bin/conda init --help for a comprehensive list of supported shells.
Word of Caution
I'd recommend running the above command with a --dry-run|-d flag and a verbosity (-v) flag, in order to see exactly what it would do. If you don't already have a Conda-managed section in your shell run commands file (e.g., .bashrc), then this should appear like a straight-forward insertion of some new lines. If it isn't such a straightforward insertion, I'd recommend clearing any previous Conda sections from $PATH and the relevant shell initialization files (e.g., bashrc) first.
Again, for those interested, there is an in-depth look at how it works, in this answer.
Potential Automated Cleanup
Conda v4.6.9 introduced a --reverse flag that automates removing the changes that are inserted by conda init.
If you have installed Anaconda but are not able to load the correct versions of python and ipython, or if you see conda: command not found when trying to use conda, this may be an issue with your PATH environment variable. At the prompt, type:
export PATH=~/anaconda/bin:$PATH
For this example, it is assumed that Anaconda is installed in the default ~/anaconda location.
This info is current as of today, August 10, 2016. Here are the exact steps I took to fix this using methods posted above. I did not see anyone post: export PATH=$PATH:$HOME/anaconda/bin (you need to add export to the beginning of the line).
Here it is, step-by-step:
For anyone running into the same problem while using oh-my-zsh, you need to do the following:
-Open your .zshrc in your terminal. I am using iTerm 2 and have Sublime Text 3 as my default text editor:
subl ~/.zshrc
-Once the file opens in your text editor, scroll to the very bottom and add:
export PATH=$PATH:$HOME/anaconda/bin
-Save the file, then close it.
-Close your terminal, then relaunch it.
-Once back in your terminal, type:
conda --v
You should then see the version of conda installed printed on your screen.
If you're using zsh, then after doing that, your terminal may show you zsh: command not found: rvm-prompt.
The solution is:
add alias rvm-prompt=$HOME/.rvm/bin/rvm-prompt within .zshrc file.
type source .zshrc.
Then the zsh: command not found: rvm-prompt will disappear.
My env: macOS & anaconda3
This works for me:
$ nano ~/.bash_profile
Add this:
export PATH=~/anaconda3/bin:$PATH
*The export path must match with the actual path of anaconda3 in the system.
Exit out and run:
$ source ~/.bash_profile
Then try:
$ jupyter notebook
Edit ~/.bash_profile, add this to it.
PATH=$PATH:$HOME/anaconda/bin
then run
source ~/.bash_profile
Hope can help you.
I'm on High Sierra MAC OS and just installed Anaconda3 via HomeBrew command. I had issue with running :
conda
It'd also give me:
-bash: conda: command not found
I tried running:
export PATH=~/anaconda3/bin:$PATH
but it needs ENTIRE path. so here are the correct steps:
$ nano ~/.bash_profile
Now export the ENTIRE path, in my case it was:
export PATH=/usr/local/anaconda3/bin:$PATH
Exit out and run:
$ source ~/.bash_profile
Then try:
$ conda
it'll output:
$ conda --version
conda 4.4.10
If you have just installed anaconda and got this error, then I think you forgot to run this command :
source ~/.bashrc
This will enable you to make use of anaconda in terminal.
This may seems simple but many (including me) do this mistake.
If the error is still persisting, you have to verify if anaconda location is added to PATH in your system.
Once you add it, you'll be fine
If you have installed Anaconda and have definitely said yes to the:
Do you wish the installer to prepend the Anaconda install location to PATH in your /home/name/.bashrc ? [yes¦no]
question then you maybe just need to restart your terminal window. The change won't take effect until you either (1) close and reopen the current window or (2) just open and start working in a new window
Basically, open a new console window and see if it works now.
For Windows:
A recent Anaconda(version 4.4.0) changed some directories.
You can find "conda" in Anaconda3/Scripts, instead of Anaconda3/bin.
Mostly it is because when we install Anaconda in the end it adds the anaconda path to PATH variable in .bashrc file.
So we just need to restart the terminal or just do
source ~/.bashrc
if still it don't work then follow this commands.
cat >> ~/.bashrc
paste the below command for anaconda3
export PATH=~/anaconda3/bin:$PATH
hit Enter then ctrl+d
source ~/.bashrc
Using the export PATH= strategy with conda activate will result into a warning message for conda>=4.4.
The recommended way (see this release note, and this post) to do is to remove the export PATH=... line and add instead in your .bashrc (.bash_profile) the line:
. ~/anaconda2/etc/profile.d/conda.sh
Furthermore, if you want conda to be activated by default, add:
conda activate
If you have installed anaconda, but if you are not able to execute conda command from terminal, it means the path is not probably set, try :
export PATH=~/anaconda/bin:$PATH
See this link.
I also was facing the same issue ,this might be the simplest possible solution
source anaconda/bin/activate
for anaconda2 use
source anaconda2/bin/activate
depending on the name of the directory , then execute your command i.e. conda --create
Run
cat ~/.bash_profile
to check if anaconda is there. If not you should add its path there.
If conda is there copy the entire row that you see the Anaconda there from "export" to the end of line.
like this:
export PATH=~/anaconda3/bin:$PATH
Run this in your terminal.
Then run
conda --version
to see if it is exported and running!
System: macOS
I installed Anaconda first, and everything worked well. Then I installed iTerm2 and when I typed Python iTerm2 gave me the default Python. Here is how to find your Anaconda Python back:
Open your .zshrc file. For example, vim ~/.zshrc
Then add export PATH="/Users/yourusername/anaconda2/bin:$PATH" to the last line of the file. Note that you need to replace the yourusername to your user name and make sure you have anaconda2. An easy way is to copy this line from ~/.bash_profile
Save the file, close it, relaunch the terminal and now Anaconda Python should be back.
Answer for
Anaconda3 5.2.0 installer
macOS HighSierra
ZSH
Unfortunately, the installer puts the PATH definition only into .bash_profile, but not the .zshrc config.
Contrary to the given answers, it doesn't (necessarily) install in ~/anaconda3/, but instead in /anaconda3/.
The PATHdefintion in .zshrc must therefore be this:
...
# Anaconda3
export PATH="/anaconda3/bin:$PATH"
...
Temporary solution:
$anaconda/bin/conda create -n py33 python=3.3 anaconda
If you have just installed MiniConda or Anaconda make sure you re-run your terminal.
From this, I mean close and open your terminal and then try conda list to verify your installation
For me, this worked!!
To edit bashrc in Ubuntu
$ /usr/bin/vim ~/.bashrc
type PATH=$PATH:$HOME/anaconda3/bin
Press Esc and :wq to save bashrc file and exit vim
enter image description here
then
$ export PATH=~/anaconda3/bin:$PATH
and type $ source ~/.bashrc
Now to confirm the installation of conda type
$ conda --version
Open your terminal and type the following command to add anaconda to environment path
for anaconda 2 :
export PATH=~/anaconda2/bin:$PATH
for anaconda 3 :
export PATH=~/anaconda3/bin:$PATH
Then to check the conda version just type,
conda --version
Create an environment using the following command,
conda create --name myenv
Activate the source using,
source activate myenv
Then your anaconda IDE is ready!!!
If you installed Anaconda with Visual Studio 2017 for Windows,
conda executable is in this path or similar.
In my case path is this:
C:\Program Files (x86)\Microsoft Visual Studio\Shared\Anaconda3_64\Scripts
Steps to add it to your PATH:
On the Windows desktop, right-click My Computer.
In the pop-up menu, click Properties.
In the System Properties window, click the Advanced tab, and then click Environment Variables.
In the System Variables window, highlight Path, and click Edit.
Add your path and restart your cmd.
You will be able to execute conda
Happy coding!
I am setting up a virtual machine running Ubuntu. I have anaconda 3 installed in the "Home" folder. When I typed "conda" into the terminal I was getting the error "conda: command not found" too.
Typing the code below into the terminal worked for me...
export PATH=$PATH:$HOME/anaconda3/bin
to check it worked I typed:
conda --version
which responded with the version number.
In my case conda Path was properly set (in .bashrc) by the conda installation bash. But to make it works I had to give executable file permissions to files in bin sub folder with chmod +x *.
My system info:
conda 4.2.9
Operating System: Debian GNU/Linux 8 (jessie)
Kernel: Linux 3.16.0-4-amd64
Architecture: x86-64
You need to put Anaconda.sh to your /home/, then run it to install. After that, you can runexport PATH=~/anaconda3/bin:$PATH . Remark: do not run Anaconda.sh in download/ directily!!!!!
I had same issue. What I did, removed anaconda directory from the install location and re-install
after that it worked as I anticipated
First, check the location of anaconda, for me I installed anaconda3 at / directory which I access with /anaconda3
Then in your terminal, input export PATH="<base location>/anaconda3/bin:$PATH" for me it's export PATH="/anaconda3/bin:$PATH".
Finally, input source $/anaconda3/bin/activate. For you, just change to your location.
Now, you could try conda list to test.
Also, visit intallation guide
1. Check where you have installed Anaconda. In my case it looks like /home/nour/anaconda3/bin
Open your .bashrc file. For example $ gedit .bashrc
3. Add this export PATH = /home/nour/anaconda3/bin:$PATH line at the end of the file and save.
Reopen the terminal. Type conda --version
NOTE: Make sure path in line no. 1 and line no. 3 must be same. In my case /home/nour/anaconda3/bin .
The main point is that as of December 2018 it's Scripts not bin.
Updating $PATH in "git bash for windows"
Use one of these:
export PATH=$USERPROFILE/AppData/Local/Continuum/anaconda2/Scripts/:$PATH
export PATH=$USERPROFILE/AppData/Local/Continuum/anaconda3/Scripts/:$PATH
Updating $PATH in the windows default command line
Use one of these:
SET PATH=%USERPROFILE%\AppData\Local\Continuum\anaconda2\Scripts\;%PATH%
SET PATH=%USERPROFILE%\AppData\Local\Continuum\anaconda3\Scripts\;%PATH%
Updating $PATH in Linux
Change /app to your installation location. If you installed anaconda change Miniconda to Anaconda. Also, check for Script vs. bin,.
export PATH="/app/Miniconda/bin:$PATH"
You may need to run set -a before setting the path, I think this is important if you're setting the path in a script. For example if you have your export command in a file called set_my_path.sh, I think you'd need to do set -a; source("set_my_path.sh").
The set -a will make your changes to the path persist for your session, but they are still not permanent.
For a more permanent solution add the command to ~/.bashrc. The installers may offer to add something like this to your ~/.bashrc file, but you can do it too (or comment it out to undo it).
General Observations:
Background: I installed the 64 bit versions of Anaconda 2 and 3 recently on my Windows 10 machine following the recommended installation steps in December of 2018.
Adding conda also enables ipython, which works much better in the native Windows command line
Following the strongly recommended installation does not add conda or ipython to the path
Anaconda 3 doesn't seem to install a command prompt application, but Anaconda 2 did have a command prompt application
The /bin folder seems to have been replaced with Scripts
Poking around in the Scripts folder is interesting, maybe the Anaconda command prompt application is in there somewhere.