Torch: Difference between revisions
No edit summary |
No edit summary |
||
| (9 intermediate revisions by 3 users not shown) | |||
| Line 1: | Line 1: | ||
{{ | <languages /> | ||
[[Category:Software]] | |||
{{Outdated}} | |||
<translate> | |||
<!--T:1--> | |||
[[Category:Software]][[Category:AI and Machine Learning]] | |||
"[http://torch.ch/ Torch] is a scientific computing framework with wide support for machine learning algorithms that puts GPUs first. It is easy to use and efficient, thanks to an easy and fast scripting language, LuaJIT, and an underlying C/CUDA implementation." | "[http://torch.ch/ Torch] is a scientific computing framework with wide support for machine learning algorithms that puts GPUs first. It is easy to use and efficient, thanks to an easy and fast scripting language, LuaJIT, and an underlying C/CUDA implementation." | ||
Torch has a distant relationship to PyTorch.<ref>See https://stackoverflow.com/questions/44371560/what-is-the-relationship-between-pytorch-and-torch, https://www.quora.com/What-are-the-differences-between-Torch-and-Pytorch, and https://discuss.pytorch.org/t/torch-autograd-vs-pytorch-autograd/1671/4 for some attempts to explain the connection.</ref> PyTorch provides a [[Python]] interface to similar | <!--T:2--> | ||
Torch has a distant relationship to PyTorch.<ref>See https://stackoverflow.com/questions/44371560/what-is-the-relationship-between-pytorch-and-torch, https://www.quora.com/What-are-the-differences-between-Torch-and-Pytorch, and https://discuss.pytorch.org/t/torch-autograd-vs-pytorch-autograd/1671/4 for some attempts to explain the connection.</ref> PyTorch provides a [[Python]] interface to software with similar functionality, but PyTorch is not dependent on Torch. See [[PyTorch]] for instructions on using it. | |||
<!--T:3--> | |||
Torch depends on [[CUDA]]. In order to use Torch you must first load a CUDA module, like so: | Torch depends on [[CUDA]]. In order to use Torch you must first load a CUDA module, like so: | ||
<!--T:4--> | |||
{{Command|module load cuda torch}} | {{Command|module load cuda torch}} | ||
== Installing Lua packages == | == Installing Lua packages == <!--T:5--> | ||
Torch comes with the Lua package manager, named [https://luarocks.org/ luarocks]. | Torch comes with the Lua package manager, named [https://luarocks.org/ luarocks]. Run | ||
luarocks list | |||
to see a list of installed packages. | |||
<!--T:13--> | |||
If you need some package which does not appear on the list, use the following to install it in your own folder: | |||
<!--T:6--> | |||
{{Command|luarocks install --local --deps-mode{{=}}all <package name>}} | {{Command|luarocks install --local --deps-mode{{=}}all <package name>}} | ||
<!--T:9--> | |||
If after this installation you are having trouble finding the packages at runtime, then add the following command<ref> https://github.com/luarocks/luarocks/wiki/Using-LuaRocks#Rocks_trees_and_the_Lua_libraries_path </ref> right before running "lua your_program.lua" | |||
command: | |||
<!--T:10--> | |||
eval $(luarocks path --bin) | |||
<!--T:11--> | |||
By experience, we often find packages that do not install well with <tt>luarocks</tt>. If you have a package that is not installed in the default module and need help installing it, please contact our [[Technical support]]. | By experience, we often find packages that do not install well with <tt>luarocks</tt>. If you have a package that is not installed in the default module and need help installing it, please contact our [[Technical support]]. | ||
<!--T:12--> | |||
<references /> | <references /> | ||
</translate> | |||
Latest revision as of 15:18, 20 March 2024
This page or section contains obsolete information and some statements may not be valid. The technical documentation is currently being updated by our support team.
"Torch is a scientific computing framework with wide support for machine learning algorithms that puts GPUs first. It is easy to use and efficient, thanks to an easy and fast scripting language, LuaJIT, and an underlying C/CUDA implementation."
Torch has a distant relationship to PyTorch.[1] PyTorch provides a Python interface to software with similar functionality, but PyTorch is not dependent on Torch. See PyTorch for instructions on using it.
Torch depends on CUDA. In order to use Torch you must first load a CUDA module, like so:
[name@server ~]$ module load cuda torch
Installing Lua packages
Torch comes with the Lua package manager, named luarocks. Run
luarocks list
to see a list of installed packages.
If you need some package which does not appear on the list, use the following to install it in your own folder:
[name@server ~]$ luarocks install --local --deps-mode=all <package name>
If after this installation you are having trouble finding the packages at runtime, then add the following command[2] right before running "lua your_program.lua" command:
eval $(luarocks path --bin)
By experience, we often find packages that do not install well with luarocks. If you have a package that is not installed in the default module and need help installing it, please contact our Technical support.
- ↑ See https://stackoverflow.com/questions/44371560/what-is-the-relationship-between-pytorch-and-torch, https://www.quora.com/What-are-the-differences-between-Torch-and-Pytorch, and https://discuss.pytorch.org/t/torch-autograd-vs-pytorch-autograd/1671/4 for some attempts to explain the connection.
- ↑ https://github.com/luarocks/luarocks/wiki/Using-LuaRocks#Rocks_trees_and_the_Lua_libraries_path