2020-12-17 11:03:45 +00:00
|
|
|
Benchmark with vanilla PyTorch
|
|
|
|
|
==============================
|
|
|
|
|
|
|
|
|
|
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.
|
|
|
|
|
In average for simple MNIST CNN classifier we are only about 0.06s slower per epoch, see detail chart bellow.
|
|
|
|
|
|
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
|
|
|
|
|
|
|
|
|
|
|
|
|
Learn more about reproducible benchmarking from the `PyTorch Reproducibility Guide <https://pytorch.org/docs/stable/notes/randomness.html>_`.
|
