Torch - Revision history

Mbaney at 13:56, 23 October 2024

2024-10-23T13:56:30Z

← Older revision		Revision as of 13:56, 23 October 2024
Line 1:		Line 1:
	You can compile [http://torch.ch/ Torch] on our ~~RHEL7~~ machines. Please make sure to [[HelpDesk \| contact staff]] if you have issues, as there are some local packages that are required and could cause the install to fail.		You can compile [http://torch.ch/ Torch] on our [[RHEL \| Red Hat Enterprise Linux (RHEL)]] machines. Please make sure to [[HelpDesk \| contact staff]] if you have issues, as there are some local packages that are required and could cause the install to fail.

	'''Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch ~~and that certain~~ newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.'''		'''Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch. Certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.'''

	First load the following software into your environment with GNU [[Modules]].		First load the following software into your environment with GNU [[Modules]].

Mbaney at 18:11, 2 March 2022

2022-03-02T18:11:02Z

← Older revision		Revision as of 18:11, 2 March 2022
Line 1:		Line 1:
	You can compile [http://torch.ch/ Torch] on our RHEL7 machines. Please make sure to ~~ask~~ staff~~@umiacs.umd.edu~~ if have issues as there are some local packages that are required and could cause the install to fail.		You can compile [http://torch.ch/ Torch] on our RHEL7 machines. Please make sure to [[HelpDesk \| contact staff]] if you have issues, as there are some local packages that are required and could cause the install to fail.

	'''Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch and that certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.'''		'''Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch and that certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.'''

Derek at 21:52, 7 March 2018

2018-03-07T21:52:25Z

← Older revision		Revision as of 21:52, 7 March 2018
Line 3:		Line 3:
	'''Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch and that certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.'''		'''Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch and that certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.'''

	First load the following software into your environment with GNU [Modules].		First load the following software into your environment with GNU [[Modules]].
	<pre>		<pre>
	module add cuda/8.0.61 cudnn zeromq nodejs sox GraphicsMagick		module add cuda/8.0.61 cudnn zeromq nodejs sox GraphicsMagick

Derek at 21:52, 7 March 2018

2018-03-07T21:52:04Z

← Older revision		Revision as of 21:52, 7 March 2018
Line 1:		Line 1:
	You can compile [http://torch.ch/ Torch] on our RHEL7 machines. Please make sure to ask staff@umiacs.umd.edu if have issues as there are some local packages that are required and could cause the install to fail.		You can compile [http://torch.ch/ Torch] on our RHEL7 machines. Please make sure to ask staff@umiacs.umd.edu if have issues as there are some local packages that are required and could cause the install to fail.

	Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch and that certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.		'''Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch and that certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.'''

	First load the following software into your environment with GNU [Modules].		First load the following software into your environment with GNU [Modules].

Derek at 21:51, 7 March 2018

2018-03-07T21:51:50Z

← Older revision		Revision as of 21:51, 7 March 2018
Line 3:		Line 3:
	Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch and that certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.		Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch and that certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.

			First load the following software into your environment with GNU [Modules].
	<pre>		<pre>
	module add cuda/8.0.61 cudnn zeromq nodejs sox GraphicsMagick		module add cuda/8.0.61 cudnn zeromq nodejs sox GraphicsMagick
			</pre>

			Next we are going to create a python environment and active it (this assumes you are using the bash shell)
			<pre>
	virtualenv env		virtualenv env
	source env/bin/activate		source env/bin/activate
			</pre>

			Then we are going to upgrade the python pip package manager and install two more dependencies
			<pre>
	pip install --upgrade pip		pip install --upgrade pip
	pip install pyzmq ipython		pip install pyzmq ipython
			</pre>

			Next clone the repository from Torch ensuring you do it with the recursive option to get all its sub dependencies.
			<pre>
	git clone https://github.com/torch/distro.git torch --recursive		git clone https://github.com/torch/distro.git torch --recursive
			</pre>

			Finally build Torch, this will take awhile.
			<pre>
	cd torch && ./install.sh		cd torch && ./install.sh
			</pre>

			At the end it will give you example of how to activate your Torch environment here is an example.
			<pre>
	source install/bin/torch-activate		source install/bin/torch-activate
	</pre>		</pre>

Derek at 21:47, 7 March 2018

2018-03-07T21:47:55Z

@@ Line 10: / Line 10: @@
 pip install pyzmq ipython
 git clone https://github.com/torch/distro.git torch --recursive
 cd torch && ./install.sh
 </pre>

Derek: Created page with "You can compile [http://torch.ch/ Torch] on our RHEL7 machines. Please make sure to ask staff@umiacs.umd.edu if have issues as there are some local packages that are required..."

2018-03-07T21:45:05Z

Created page with "You can compile [http://torch.ch/ Torch] on our RHEL7 machines. Please make sure to ask staff@umiacs.umd.edu if have issues as there are some local packages that are required..."

New page

You can compile [http://torch.ch/ Torch] on our RHEL7 machines. Please make sure to ask staff@umiacs.umd.edu if have issues as there are some local packages that are required and could cause the install to fail.

Note you have to use Cuda 8 or Cuda 7.5 when compiling Torch and that certain newer GPU cards are not supported under Cuda 7.5. We will use Cuda 8 in this example.

<pre>
module add cuda/8.0.61 cudnn zeromq nodejs sox GraphicsMagick
virtualenv env
source env/bin/activate
pip install --upgrade pip
pip install pyzmq ipython
git clone https://github.com/torch/distro.git torch --recursive
cd torch && ./install.sh
</pre>