Skip to content

[minibench] Drop outliers from benchmark result #8919

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 4 commits into from
Mar 4, 2025
Merged

[minibench] Drop outliers from benchmark result #8919

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
646f6cc
[minibench] Drop outliers from benchmark result
kirklandsign Mar 4, 2025
62d5b73
Merge branch 'main' into kirk-use-trimmean
kirklandsign Mar 4, 2025
a622dc1
format
kirklandsign Mar 4, 2025
ce0902f
Merge branch 'main' into kirk-use-trimmean
kirklandsign Mar 4, 2025
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
10 changes: 9 additions & 1 deletion ...enchmark/android/benchmark/app/src/main/java/org/pytorch/minibench/BenchmarkActivity.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
import java.util.List;
import java.util.stream.Collectors;
import org.pytorch.executorch.Module;
Expand Down Expand Up @@ -80,11 +81,18 @@ protected void onPostExecute(Void aVoid) {
final List<BenchmarkMetric> results = new ArrayList<>();
// The list of metrics we have atm includes:
// Avg inference latency after N iterations
// Currently the result has large variance from outliers, so only use
// 80% samples in the middle (trimmean 0.2)
Collections.sort(stats.latency);
int resultSize = stats.latency.size();
List<Double> usedLatencyResults =
stats.latency.subList(resultSize / 10, resultSize * 9 / 10);

results.add(
new BenchmarkMetric(
benchmarkModel,
"avg_inference_latency(ms)",
stats.latency.stream().mapToDouble(l -> l).average().orElse(0.0f),
usedLatencyResults.stream().mapToDouble(l -> l).average().orElse(0.0f),
0.0f));
// Model load time
results.add(
Expand Down
Loading