Skip to content
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

Jump to bottom

MLPerf inference v4.1 Postmortem item: Rename <system_desc_id>_<implementation_id>_<scenario>.json to model-info.json #182

Merged
merged 5 commits into from
Nov 19, 2024
Merged

MLPerf inference v4.1 Postmortem item: Rename <system_desc_id>_<implementation_id>_<scenario>.json to model-info.json #182

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
301707f
Update submission_rules.adoc
arjunsuresh Sep 10, 2024
3d399ab
Merge branch 'master' into patch-6
arjunsuresh Sep 27, 2024
7f91adc
Merge branch 'master' into patch-6
arjunsuresh Oct 25, 2024
49454af
Update submission_rules.adoc | Support impl in model-info, added opti…
arjunsuresh Oct 25, 2024
92a374a
Merge branch 'master' into patch-6
mrasquinha-g Nov 19, 2024
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: 5 additions & 4 deletions submission_rules.adoc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -373,7 +373,7 @@ HPC training submissions follow the above Training directory structure except fo
*** <system_desc_id>/
**** <benchmark>/
***** <scenario>
****** <system_desc_id>_<implementation_id>_<scenario>.json
****** model-info.json or model-info_<implementation_id>.json
****** README.md
****** user.conf
****** mlperf.conf
Expand Down Expand Up @@ -427,7 +427,7 @@ Here is the list of mandatory files for all submissions in any division/category
* actual models if the models are not deterministically generated
* READMEs to enable users to replicate performance results
* code which interfaces with the loadgen
* <system_desc_id>_<implementation_id>_<scenario>.json
* model-info.json or model-info_<implementation_id>.json
* <system_desc_id>.json

For some models mlperf_log_accuracy.json can get very large. Because of this we truncate mlperf_log_accuracy.json in submissions
Expand Down Expand Up @@ -511,9 +511,9 @@ In the Network division for the inference datacenter the file <system_desc_id>.j



### <system_desc_id>_<implementation_id>_<scenario>.json metadata
### model-info_<implementation_id>.json metadata

The file <system_desc_id>_<implementation_id>.json should contain metadata describing use of the specified implementation on the specified system.
The file `model-info.json` or `model-info_<implementation_id>.json` should contain metadata describing use of the specified implementation on the specified system.


|===
Expand All @@ -523,6 +523,7 @@ The file <system_desc_id>_<implementation_id>.json should contain metadata descr
| Weight data type(s) | Yes | fp32 | bf16
| Input data type(s) | Yes | fp32 | bf16
| Retraining | Yes | No | Yes (URL_to_writeup)
| Notes | No | Run on only one GPU | DLA cores are used for the run
|===


Expand Down