All questions related to Real-Time Camera Scene Detection Challenge can be asked in this thread.
Is there any runtime limit for this competition?
If I use a mobile device to test, can I only use Samsung Galaxy S10 (Exynos) in this competition?
Does input resolution of tflite should also be the size of dataset that is given in this competition?
Do you have any specific rules of the ranking? Such as what is the weight of inference speed in the grade?
Thanks for pointing out the error, my submission was successful.I've checked the latest submission and I think that your .txt files are not in the right format.
Attached is an example of .txt files that you could check.
Thanks for resolving the issue!I've checked the evaluation script and re-run your entry.
You've got 94%!
If you have any other questions please let me know.
There were connection problems with the mobile device. The service was restarted.Is http://lightspeed.difficu.lt:60002/ experiencing some problem? It shows error even when I submit an old model that ran successfully before.
Thank you for your response, the submission was successful, however, the output accuracy is zero. How did you solve the same problem earlier? Thanks.you need to have your files in the root of the zip archive, without any folders or subfolders
can you check validation datset number 189.jpg? it seems have 2 dimensions
"Test on bionic" server is not working. The status always be "In progress" and I can't click the "Check Model" button
CPU, GPU, and NNAPI, which mode should be used when running on my device? It is worth noting that some type of models run much faster in CPU than in NNAPI
so which mode should be used when running on my device?The server is working from Sunday without any problems, just wait till the evaluation of the previously submitted model is finished.
This completely depends on the phone / SoC model you are using - many of them are not able to accelerate quantized models efficiently. Please check this ranking table and paper for more details.
so which mode should be used when running on my device?
Can I use extra training data for this competition?All questions related to Real-Time Camera Scene Detection Challenge can be asked in this thread.
Can I use extra training data for this competition?
Is private extra training data also OK?Sure, you can use extra training data, this should be just indicated in your final challenge report.
Is private extra training data also OK?
What is the form of submission during the test phase, or is it the same as in the validation phase, such as downloading test images, offline testing, submission of test results file (TXT) and model files(TFLite)?
do we all need to register for the challenge on CodaLab, or only one of us?
I am submitting the zip file named as submission.zip
"The runtime of your final model will be evaluated on the Apple Bionic platform". Which one among CPU, GPU, DSP and NPU will the final runtime test of the tflite model run on?All questions related to Real-Time Camera Scene Detection Challenge can be asked in this thread.
Also which Bionic SoC will be used?"The runtime of your final model will be evaluated on the Apple Bionic platform". Which one among CPU, GPU, DSP and NPU will the final runtime test of the tflite model run on?
Also how the final inference time will be calculated?
On real device?
Using lightspeed?
Which one among CPU, GPU, DSP and NPU will the final runtime test of the tflite model run on?
which Bionic SoC will be used?
guess as long as all contestants are evaluated on a same platform/device/build settings, right?