Skip to content

Commit e8fe852

Browse files
committed
Add benchmark scripts
Fixes #17 A continuation of #81 -- see discussion there.
1 parent c8d20b6 commit e8fe852

File tree

4 files changed

+431
-1
lines changed

4 files changed

+431
-1
lines changed

benchmark/README.rst

Lines changed: 23 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,23 @@
1+
``benchmark``: performance sanity checking
2+
==========================================
3+
4+
This provides a basic benchmarking script intended to catch major performance regressions (or improvements!). It emulates a "training"-style workload of computing energy and forces for a number of similar systems with pre-computed neighborlists and settings. In particular, we run PME forward and backward calculations for supercells of cubic CsCl crystals, in different sizes. Results are stored as a time-stamped ``.yaml`` file, together with some basic system and version information (which is just the output of ``git`` for now).
5+
6+
7+
Usage
8+
-----
9+
10+
First, make sure to install ``torch-pme`` with the right dependencies in a fresh enviroment:
11+
12+
.. code-block:: bash
13+
14+
pip install .[metatensor,examples]
15+
16+
Then, run
17+
18+
.. code-block:: bash
19+
20+
python run.py
21+
22+
23+
Please include the output ``.yaml`` file in any major pull request. We recommend using a H100 as benchmark GPU (internally, we use the ``kuma`` cluster).

benchmark/geometry.in

Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,11 @@
1+
#===============================================================================
2+
# Created using the Atomic Simulation Environment (ASE)
3+
4+
# Fri Sep 20 14:44:08 2024
5+
6+
#=======================================================
7+
lattice_vector 1.0000000000000000 0.0000000000000000 0.0000000000000000
8+
lattice_vector 0.0000000000000000 1.0000000000000000 0.0000000000000000
9+
lattice_vector 0.0000000000000000 0.0000000000000000 1.0000000000000000
10+
atom 0.0000000000000000 0.0000000000000000 0.0000000000000000 Cs
11+
atom 0.5000000000000000 0.5000000000000000 0.5000000000000000 Cl

0 commit comments

Comments
 (0)