The benchmark binary support multiple batches in one run (#15443)
authorFei Sun <feisun@fb.com>
Fri, 21 Dec 2018 16:39:05 +0000 (08:39 -0800)
committerFacebook Github Bot <facebook-github-bot@users.noreply.github.com>
Fri, 21 Dec 2018 16:45:41 +0000 (08:45 -0800)
commit2b23ba8ef089392ab0d2d62cbe0ce53ba879b3bf
tree914b68169068b99ba79421f99a42312d910d6b5e
parent433db13b48ea4ac56847ac5a664db9abfdb8c09a
The benchmark binary support multiple batches in one run (#15443)

Summary:
It is sometimes beneficial to run multiple batches in one benchmark and check the aggregated results.

This PR enables this functionality.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/15443

Reviewed By: llyfacebook

Differential Revision: D13531129

Pulled By: sf-wind

fbshipit-source-id: 553a762a5cbadf5a3d9fd6af767ae34899bc1aa2
binaries/benchmark_helper.cc
binaries/benchmark_helper.h
binaries/convert_image_to_tensor.cc