Support for custom metrics reported in the Benchmarks#735

Merged

safern merged 13 commits intodotnet:masterfrom

adamsitnik:benchmarksPolishing

Aug 30, 2018

Commits on Aug 26, 2018

simplify and cleanup the code, remove dead code
adamsitnik
committed
use `Target` to specify that given setup method should be executed for selected benchmarks, not all
adamsitnik
committed
consume the result of Predict to make sure it does not get dead-code eliminated
adamsitnik
committed
reference input files from .csproj and copy them to output dir, don't rely on hardcoded folder hierarchy
adamsitnik
committed
every ML.NET benchmark allocates a lot of memory and should be executed in a dedicated process
adamsitnik
committed
make it possible for every type to report different metrics
adamsitnik
committed
enforce current culture as "en-us" because the input data files use dot as decimal separator (and it fails for cultures with ",")
adamsitnik
committed
for our time consuming benchmarks 1 warmup iteration is enough
adamsitnik
committed
workaround for the auto-generated code to avoid name coflict for Microsoft.ML.Runtime.IHost and BenchmarkDotNet.Engines.IHost..
adamsitnik
committed
add comment about why we need a custom toolchain
adamsitnik
committed
update BDN version to allow benchmarking with CoreRun
adamsitnik
committed

Commits on Aug 29, 2018

code review fix: spacing
adamsitnik
committed

Commits on Aug 30, 2018

Merge remote-tracking branch 'upstream/master' into benchmarksPolishing
adamsitnik
committed