rsnt_translations
56,420
edits
Large Scale Machine Learning (Big Data): Difference between revisions
Jump to navigation
Jump to search
Large Scale Machine Learning (Big Data) (view source)
Revision as of 17:34, 16 March 2022
, 2 years agoMarked this version for translation
(Marked this version for translation) |
|||
| Line 137: | Line 137: | ||
[https://snapml.readthedocs.io/en/latest/ Snap ML] is a closed-source Machine Learning library being developed by IBM. <code>Snap ML</code> currently supports a number of classical machine learning models and scales gracefully to data sets with billions of examples and/or features. It offers distributed training, GPU acceleration and supports sparse data structures. It features an API very similar to <code>scikit-learn</code> and can be used as a replacement for that library when dealing with massive datasets. | [https://snapml.readthedocs.io/en/latest/ Snap ML] is a closed-source Machine Learning library being developed by IBM. <code>Snap ML</code> currently supports a number of classical machine learning models and scales gracefully to data sets with billions of examples and/or features. It offers distributed training, GPU acceleration and supports sparse data structures. It features an API very similar to <code>scikit-learn</code> and can be used as a replacement for that library when dealing with massive datasets. | ||
== Installation == | == Installation == <!--T:69--> | ||
===Latest available wheels=== | ===Latest available wheels=== <!--T:70--> | ||
To see the latest version of Snap ML that we have built: | To see the latest version of Snap ML that we have built: | ||
{{Command|avail_wheels "snapml"}} | {{Command|avail_wheels "snapml"}} | ||
For more information, see [[Python#Available_wheels |Available wheels]]. | For more information, see [[Python#Available_wheels |Available wheels]]. | ||
===Installing the Compute Canada wheel=== | ===Installing the Compute Canada wheel=== <!--T:71--> | ||
<!--T:72--> | |||
The preferred option is to install it using the Python [https://pythonwheels.com/ wheel] as follows: | The preferred option is to install it using the Python [https://pythonwheels.com/ wheel] as follows: | ||
:1. Load a Python [[Utiliser_des_modules/en#Sub-command_load|module]], thus <tt>module load python</tt> | :1. Load a Python [[Utiliser_des_modules/en#Sub-command_load|module]], thus <tt>module load python</tt> | ||
| Line 151: | Line 152: | ||
:3. Install SnapML in the virtual environment with <code>pip install</code>. | :3. Install SnapML in the virtual environment with <code>pip install</code>. | ||
<!--T:73--> | |||
:{{Command|prompt=(venv) [name@server ~]|pip install --no-index snapml }} | :{{Command|prompt=(venv) [name@server ~]|pip install --no-index snapml }} | ||