-
Notifications
You must be signed in to change notification settings - Fork 169
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.
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
Tweak tuning database plot and comparison scripts #2883
Conversation
🟩 CI finished in 5h 06m: Pass: 100%/400 | Total: 8d 04h | Avg: 29m 31s | Max: 1h 25m | Hits: 11%/25894
|
Project | |
---|---|
+/- | CCCL Infrastructure |
libcu++ | |
CUB | |
Thrust | |
CUDA Experimental | |
python | |
CCCL C Parallel Library | |
Catch2Helper |
Modifications in project or dependencies?
Project | |
---|---|
+/- | CCCL Infrastructure |
+/- | libcu++ |
+/- | CUB |
+/- | Thrust |
+/- | CUDA Experimental |
+/- | python |
+/- | CCCL C Parallel Library |
+/- | Catch2Helper |
🏃 Runner counts (total jobs: 400)
# | Runner |
---|---|
326 | linux-amd64-cpu16 |
31 | linux-amd64-gpu-v100-latest-1 |
28 | linux-arm64-cpu16 |
15 | windows-amd64-cpu16 |
def print_speedup(medians): | ||
m = medians.groupby(['alg', 'hue'], sort=False).mean() | ||
m['speedup'] = (m['bw'] / m.groupby(['alg'])['bw'].transform('first')) | ||
print('# Speedups:') | ||
print() | ||
print(m.drop(columns='bw').sort_values(by='speedup', ascending=False).to_markdown()) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This feels a bit misplaced here, but it's great to have the speedup's from the plot as a copyable table as well., since compare.py
works on a much more detailed level.
* Mark compare as executable * Plot and compare each sub bench * Avoid confusing use of 'bw' * Don't filter the runs
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
The scripts are definitely a bit hacky, but I need those changes regularly, so let's proceed and get them merged. |
🟩 CI finished in 5h 15m: Pass: 100%/402 | Total: 6d 20h | Avg: 24m 30s | Max: 1h 24m | Hits: 68%/22121
|
Project | |
---|---|
+/- | CCCL Infrastructure |
libcu++ | |
CUB | |
Thrust | |
CUDA Experimental | |
python | |
CCCL C Parallel Library | |
Catch2Helper |
Modifications in project or dependencies?
Project | |
---|---|
+/- | CCCL Infrastructure |
+/- | libcu++ |
+/- | CUB |
+/- | Thrust |
+/- | CUDA Experimental |
+/- | python |
+/- | CCCL C Parallel Library |
+/- | Catch2Helper |
🏃 Runner counts (total jobs: 402)
# | Runner |
---|---|
327 | linux-amd64-cpu16 |
32 | linux-amd64-gpu-v100-latest-1 |
28 | linux-arm64-cpu16 |
15 | windows-amd64-cpu16 |
Here are some tweaks I made to @gevtushenko tuning database plot and comparison scripts. It fixes a few bugs and changes the behavior from plotting aggregated results across sub benchmarks (e.g. aggregating transform.add and transform.triad) to plotting each sub benchmark separately.