Anh
Anh
Another solution would be using GitHub Release if each file is under 2 GB. The SNUSR dataset was released this way: https://github.com/mlcommons/mobile_models/releases
Waiting until https://github.com/mlcommons/mobile_app_open/pull/707 is merged so we save time and bandwidth downloading dataset.
We can either rename `benchmark -> task` or `task -> benchmark` since the 2 terms are used interchangeable here. Actually, renaming `task -> benchmark` would be easier since it does...
One issue with using a user editable configuration file is that if we want to update the path to the default task file, it will not work if the users...
> if we want to update the path to the default task file, it will not work if the users still have the old configuration file It seems to be...
**Quick summary:** We started with Firebase Database to collect benchmark results. But Firebase Database has limited options to filter the results. So after some discussions and considerations, we decided to...
The data collected is described here: https://github.com/mlcommons/mobile_app_open/blob/master/docs/result-spec.md
I can think of the following cases: 1. Add a new field: Adding new fields is a very common scenario. All DB system will do it without any issue. I...
@RSMNYS, please create a PR for the changes you'd like to implement. Once that's done, I'll replace the URL with the one from MLCommons's cloud bucket.
@RSMNYS You can use custom_setting for model file name, please see https://github.com/mlcommons/mobile_app_open/pull/924