Skip to content
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.

Benchmark doc fix #15769

Merged
merged 8 commits into from
Aug 20, 2019
Merged

Benchmark doc fix #15769

Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
60c4dc7
Update pre-req for opperf
ChaiBapchya Aug 6, 2019
5de11b8
Update README.md
ChaiBapchya Aug 6, 2019
d7b66c4
Merge branch 'master' into benchmark_doc_fix
ChaiBapchya Aug 13, 2019
1db2196
correct command to import binary broadcast
ChaiBapchya Aug 13, 2019
c1999cf
no such op called nd.sub, it is nd.subtract
ChaiBapchya Aug 13, 2019
964f3dd
Trigger notification
ChaiBapchya Aug 14, 2019
0bfd64e
Trigger notification
ChaiBapchya Aug 15, 2019
09d0c89
Merge branch 'master' into benchmark_doc_fix
ChaiBapchya Aug 15, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 6 additions & 3 deletions benchmark/opperf/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,7 +47,10 @@ Hence, in this utility, we will build the functionality to allow users and devel

## Prerequisites

Make sure to build the flavor of MXNet, for example - with/without MKL, with CUDA 9 or 10.1 etc., on which you would like to measure operator performance. Finally, you need to add path to your cloned MXNet repository to the PYTHONPATH.
Provided you have MXNet installed (any version >= 1.5.1), all you need to use opperf utility is to add path to your cloned MXNet repository to the PYTHONPATH.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It would be great to give a one line example.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In the note I have added link to how to install MXNet (page) instead of giving just 1 of the 10 different ways to install mxnet


Note:
To install MXNet, refer [Installing MXNet page](https://mxnet.incubator.apache.org/versions/master/install/index.html)

```
export PYTHONPATH=$PYTHONPATH:/path/to/incubator-mxnet/
Expand Down Expand Up @@ -75,7 +78,7 @@ For example, you want to run benchmarks for all NDArray Broadcast Binary Operato

```
#!/usr/bin/python
from benchmark.opperf.nd_operations.binary_broadcast_operators import run_mx_binary_broadcast_operators_benchmarks
from benchmark.opperf.nd_operations.binary_operators import run_mx_binary_broadcast_operators_benchmarks

# Run all Binary Broadcast operations benchmarks with default input values
print(run_mx_binary_broadcast_operators_benchmarks())
Expand Down Expand Up @@ -136,7 +139,7 @@ from mxnet import nd

from benchmark.opperf.utils.benchmark_utils import run_performance_test

add_res = run_performance_test([nd.add, nd.sub], run_backward=True, dtype='float32', ctx=mx.cpu(),
add_res = run_performance_test([nd.add, nd.subtract], run_backward=True, dtype='float32', ctx=mx.cpu(),
inputs=[{"lhs": (1024, 1024),
"rhs": (1024, 1024)}],
warmup=10, runs=25)
Expand Down