GIZ NLP Agricultural Keyword Spotter
$7,000 USD
Classify audio utterances in Luganda and English from Uganda
673 data scientists enrolled, 231 on the leaderboard
AgricultureClassificationAudioNLPUnstructuredSDG2
Uganda
11 September—29 November 23:59
Ends in 65 hours
Test set
published 9 Nov 2020, 14:39

Hi everyone!

I'm a little bit confused about the test set - or the lack thereof. The predictions of which audio files should be submitted? The files in the folder "audiofiles"? But don't we have already the labels for them (and so make it possible to cheat)?

Unless I'm misunderstanding something...

Thank you so much in advance!

Hi!

The filenames for the audio files that make up the test set are in the SampleSubmission.csv file. The audio_files.zip file contains both audio files for training (whose FileNames are in Train.csv) and also the audio files for which you are expected to make predictions for. Hence your submission should only be comprised of audio files which are in SampleSubmission.csv and not necessarily all files within audio_files.zip.

Happy coding!