Difference between revisions of "SW:Magenta"
(→Magenta via a Python virtual environment) |
(→Magenta via a Python virtual environment) |
||
| Line 8: | Line 8: | ||
A lot of what is described below is probably easily adapted to [[SW:Anaconda | Anaconda]], but we are going to use a plain Python for now | A lot of what is described below is probably easily adapted to [[SW:Anaconda | Anaconda]], but we are going to use a plain Python for now | ||
==== Magenta via a Python virtual environment ==== | ==== Magenta via a Python virtual environment ==== | ||
| + | In order to get the most out of Magenta you'll need the following modules: | ||
| + | # '''cuDNN/7.0.5-CUDA-9.0.176''' - the version of Magenta installed by pip strictly requires CUDA 9.0 and cuDNN 7.0 | ||
| + | # an "oldish" Python. Python 3.6 is too new. At the moment we recommend the Python 3.5s from the [[SW:Toolchains | 2017A toolchain]]. | ||
test | test | ||
<pre> | <pre> | ||
this is code | this is code | ||
</pre> | </pre> | ||
Revision as of 05:25, 6 December 2018
Contents
Magenta
"Magenta is distributed as an open source Python library, powered by TensorFlow. This library includes utilities for manipulating source data (primarily music and images), using this data to train machine learning models, and finally generating new content from these models."
Magenta on HPRC clusters
Magenta module
Ideally we would rebuild everything needed for Magenta to optimize for our clusters but this is a significant and less than straightforward task. So for now we recommend the Python solution below.
Magenta via an Anaconda virtual environment
A lot of what is described below is probably easily adapted to Anaconda, but we are going to use a plain Python for now
Magenta via a Python virtual environment
In order to get the most out of Magenta you'll need the following modules:
- cuDNN/7.0.5-CUDA-9.0.176 - the version of Magenta installed by pip strictly requires CUDA 9.0 and cuDNN 7.0
- an "oldish" Python. Python 3.6 is too new. At the moment we recommend the Python 3.5s from the 2017A toolchain.
test
this is code
