History

hailin c8e8c1e9ff .		2025-09-20 16:09:34 +08:00
..
distributed/ddp	.	2025-09-20 16:09:34 +08:00
dynamo	.	2025-09-20 16:09:34 +08:00
fastrnns	.	2025-09-20 16:09:34 +08:00
framework_overhead_benchmark	.	2025-09-20 16:09:34 +08:00
functional_autograd_benchmark	.	2025-09-20 16:09:34 +08:00
fuser	.	2025-09-20 16:09:34 +08:00
gpt_fast	.	2025-09-20 16:09:34 +08:00
inductor_backends	.	2025-09-20 16:09:34 +08:00
inference	.	2025-09-20 16:09:34 +08:00
instruction_counts	.	2025-09-20 16:09:34 +08:00
nested	.	2025-09-20 16:09:34 +08:00
operator_benchmark	.	2025-09-20 16:09:34 +08:00
overrides_benchmark	.	2025-09-20 16:09:34 +08:00
profiler_benchmark	.	2025-09-20 16:09:34 +08:00
record_function_benchmark	.	2025-09-20 16:09:34 +08:00
serialization	.	2025-09-20 16:09:34 +08:00
sparse	.	2025-09-20 16:09:34 +08:00
static_runtime	.	2025-09-20 16:09:34 +08:00
tensorexpr	.	2025-09-20 16:09:34 +08:00
transformer	.	2025-09-20 16:09:34 +08:00
README.md	.	2025-09-20 16:09:34 +08:00
compare-fastrnn-results.py	.	2025-09-20 16:09:34 +08:00
compare.sh	.	2025-09-20 16:09:34 +08:00
upload_scribe.py	.	2025-09-20 16:09:34 +08:00

README.md

PyTorch Benchmarks

This folder contains scripts that produce reproducible timings of various PyTorch features.

It also provides mechanisms to compare PyTorch with other frameworks.

Setup environment

Make sure you're on a machine with CUDA, torchvision, and pytorch installed. Install in the following order:

# Install torchvision. It comes with the pytorch stable release binary
pip3 install torch torchvision

# Install the latest pytorch master from source.
# It should supersede the installation from the release binary.
cd $PYTORCH_HOME
python setup.py build develop

# Check the pytorch installation version
python -c "import torch; print(torch.__version__)"

Benchmark List

Please refer to each subfolder to discover each benchmark suite. Links are provided where descriptions exist: