2022-04-19 18:15:47 +00:00
|
|
|
:orphan:
|
|
|
|
|
|
2023-04-02 17:54:59 +00:00
|
|
|
Benchmark performance vs. vanilla PyTorch
|
|
|
|
|
=========================================
|
2020-12-17 11:03:45 +00:00
|
|
|
|
|
|
|
|
In this section we set grounds for comparison between vanilla PyTorch and PT Lightning for most common scenarios.
|
|
|
|
|
|
|
|
|
|
Time comparison
|
|
|
|
|
---------------
|
|
|
|
|
|
|
|
|
|
We have set regular benchmarking against PyTorch vanilla training loop on with RNN and simple MNIST classifier as per of out CI.
|
2023-09-26 15:54:44 +00:00
|
|
|
In average for simple MNIST CNN classifier we are only about 0.06s slower per epoch, see detail chart below.
|
2020-12-17 11:03:45 +00:00
|
|
|
|
2021-02-03 15:08:19 +00:00
|
|
|
.. figure:: ../_static/images/benchmarks/figure-parity-times.png
|
2020-12-17 11:03:45 +00:00
|
|
|
:alt: Speed parity to vanilla PT, created on 2020-12-16
|
|
|
|
|
:width: 500
|
2021-08-27 07:59:15 +00:00
|
|
|
|
|
|
|
|
|
2022-03-28 18:21:11 +00:00
|
|
|
Learn more about reproducible benchmarking from the `PyTorch Reproducibility Guide <https://pytorch.org/docs/stable/notes/randomness.html>`__.
|
2023-04-02 17:54:59 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
----
|
|
|
|
|
|
|
|
|
|
Find performance bottlenecks
|
|
|
|
|
=============================
|
|
|
|
|
|
|
|
|
|
.. raw:: html
|
|
|
|
|
|
|
|
|
|
<div class="display-card-container">
|
|
|
|
|
<div class="row">
|
|
|
|
|
|
|
|
|
|
.. Add callout items below this line
|
|
|
|
|
|
|
|
|
|
.. displayitem::
|
|
|
|
|
:header: Find bottlenecks in your models
|
|
|
|
|
:description: Benchmark your own Lightning models
|
|
|
|
|
:button_link: ../tuning/profiler.html
|
|
|
|
|
:col_css: col-md-3
|
|
|
|
|
:height: 180
|
|
|
|
|
:tag: basic
|
|
|
|
|
|
|
|
|
|
.. raw:: html
|
|
|
|
|
|
|
|
|
|
</div>
|
|
|
|
|
</div>
|
