This took me an excruciatingly long time for me to figure out so I decided to save people the trouble.
Trying to install torch on an ec2 instance with Amazon Linux(redhat) failed with the following command:
git clone https://github.com/torch/distro.git ~/torch --recursive
cd ~/torch; bash install-deps;
./install.sh
It gave me the error:
OpenBLAS Failed to compile
Turns out that you just run the install-deps script and ignore the failure, then run the following line of code:
export TORCH_NVCC_FLAGS="-D__CUDA_NO_HALF_OPERATORS__"
and follow with the install.sh script as usual:
./install.sh
Source Here
Related
I am trying to install from a GitHub repo with bash code (the pip installation doesn't work).
I'm trying the following code (s. here for reference):
https://github.com/vi3k6i5/GuidedLDA
cd GuidedLDA
sh build_dist.sh
python setup.py sdist
pip install -e .
which I adapted for Colab to
!https://github.com/vi3k6i5/GuidedLDA
!cd GuidedLDA
!sh build_dist.sh
!python setup.py sdist
!pip install -e .
However, already the first line throws an Error: /bin/bash: https://github.com/vi3k6i5/GuidedLDA: No such file or directory
I am very new to bash, so apologies for rookie mistakes! I tried also
!"https://github.com/vi3k6i5/GuidedLDA"
!github.com/vi3k6i5/GuidedLDA
!"github.com/vi3k6i5/GuidedLDA"
Then I tried using xdg-open with
!xdg-open https://github.com/vi3k6i5/GuidedLDA
and
!apt-get install xdg-open
!xdg-open https://github.com/vi3k6i5/GuidedLDA
all to no avail.
Can someone point me in the right direction?
Previously I used the git throw command.
it was difficult for me.
Now I use GitHub desktop.
it is too easy to handle everything.
you can check this playlist ->
https://www.youtube.com/playlist?list=PLcL8q_TiioW0JLk03hM3cu_Nb5DRwBHa1
I am trying to write a script, which starts pgadmin4. The program starts if I run the script python3 web/pgAdmin4.py from its own folder, but that isn't as fast as running a command from $PATH...
I managed to write a shell script based on other answers from different posts, but sadly the virtual environment still doesn't work (I assume).
The shell script:
#!/bin/bash
source ~/pgadmin4/venv/bin/activate
python3 ~/pgadmin4/web/pgAdmin4.py
The output:
If you have followed these installation steps no need to activate the virtual environment. You can call the executable file pgadmin4 under bin directory that will be executed with the pgadmin4 virtual environment:
#!/bin/bash
~/pgadmin4/venv/bin/pgadmin4
The error message you provided indicates that you lack the "flask" module. Please install it through "pip install flask" or "python3 - m pip install flask".
python3 -m pip install flask
Problem Description
I'm having trouble making commits, when I try to make a commit with the command, for example:
$ git commit -m "add readme"
pre-commit not found. Install pre-commit with the command pip3 install --user pre-commit or follow the steps on official documentation: https://pre-commit.com /#install
Following the steps described in the pre-commit installation documentation I installed by the command:
$ pip install pre-commit
However when I trigger the command the following error occurs:
$ pre-commit --version
bash: pre-commit: command not found
My attempt fails
I've tried some other solutions for this but they didn't work:
Export bash
I already tried this solution described that would be to export my ~./bashrc with the command: source ~/.profile but the following error happens:
bash:/Users/pvieira/.profile: No such file or directory
Install using homebrew
Doing so the same error that occurs when installing by pip appears here.
I was able to solve the problem by simply restarting the terminal.
I solved this problem by restarting my ubuntu system
The following is what I am trying to acheive:
git clone git_project_url
cd git_project_url
./install.sh /usr/local
This library is requirement for a project and I want to install this library into the already existing install script..
So basically, I am guessing.. the idea is to clone the project into tmp..
and then install it into that specific location... and well halt if the install fails for any particular reason..
How do I automate this in a shell script.
Thanks
Write your shell script like this:
git clone git_project_url &&
cd git_project_url &&
./install.sh /usr/local
If either of the first two commands (git or cd) fails the remaining commands won't be executed.
This won't change the behaviour of ./install.sh however. If it continues running after encountering errors when you run it from the shell prompt, it will also continue to run after errors when you in this script.
I'm trying to test Rasumus Andersson's ec2-webapp but it is all scripted in Move and I'm having trouble getting Move working.
I installed it with 'npm install move' per the instructions, but when I try to run a Move script (like ec2-webapp/bin/myapp-httpd/mv), I get the error:
/usr/bin/env: move: No such file or directory
When I change the shebang in that script to a direct link to the move installation (#!/var/ec2-webapp/src/node/node_modules/move/bin).
-bash: bin/myapp-httpd.mv: /var/ec2-webapp/src/node/node_modules/move/bin: bad interpreter: Permission denied
I'm not sure how to get Move into env correctly, any suggestions?
I was just following the same tutorial by Rasumus Andersson and ran into the very same problem.
When you install npm modules, you can install them locally or globally.
http://blog.nodejs.org/2011/03/23/npm-1-0-global-vs-local-installation/
When I installed move globally, the error went away:
sudo npm -g install move