-
we have at least the following issues
1. the latency setting,
2. the TensorFlow 2.80,
3. uploading new (converted, e.g., QIT's dlc files and original (mosaic))
Let's start the merging after th…
-
Found that android apk generated using flutter is not able to load libqtibackend.so.
On debugging, found that the issue is with the latest android version 31. So reverted the targetSDKversion to 30.
…
-
I just downloaded param ssd-resnet34 1200x1200 pytorch at [table](https://github.com/mlcommons/inference/tree/master/vision/classification_and_detection) and ran it.
And result is like this. mAP:19…
-
Application currently requires each workload to have specific benchmark_id , there are no benchmark IDs for SSD-MobileDet, MobileBert and MOSAIC workloads in offline scenario. Can this be implemented …
-
Submission checklist needs to be updated for v1.0:
https://github.com/mlcommons/inference_policies/blob/master/submission_checklist.md
The rules regarding the submission checklist are still not de…
-
Environment :
Windows OS
TFLite Backend
Flutter app
Steps to reproduce :
1. git clone https://github.com/mlcommons/mobile_app_open.git
2. cd mobile_app_open
3. make flutter
4. flutter run
…
-
Print version of individual CM components to make sure that everything is updated.
-
Currently the app only shows last result. When you rerun any set of benchmarks, results of the previous run are erased.
We need to add a new UI page that will show list of previous results and allow …
-
**Describe the bug**
NaNs in results.yaml files appear as ".nan", which causes JSON Decode errors when submitting them over the REST API.
**To Reproduce**
Steps to reproduce the behavior:
includ…
-
Steps to reproduce:
1) Make sure any one of the model fails to run and results in crash (app closes).
2) Run the suite by clicking on "go".
3) Once the app crashes, open the app again to find that…