Dear participants,
\n\nThe competition has concluded, and Stage 2 has begun. Kaggle is re-running kernels on the hidden test set. The extended deadline is to let you know that the competition is still active, and the leaderboard is still being updated. You can expect final results in a couple of weeks.
\n\nWe’d like to thank everyone for your passionate participation. It has been a few intense months where both participants and organizers have learnt a lot. 881 teams and 12,869 submissions! Way more than last year’s edition of the competition. We would like to give special thanks to those who contributed in the Discussion forums sharing their knowledge, and also to those who submitted through the DCASE system. This will have an impact on the research community.
\n\nSome announcements:
\n\nIn the coming days, submissions and code may be reviewed for the winners and for some other teams.
You can update your technical reports to the DCASE Challenge until 17 June 2019 (AoE). http://dcase.community/challenge2019/submission
After all the technical reports have been updated, the organizers will begin the review of the Judges’ Award submissions.
Regardless whether you submitted a technical report or not, we encourage you all to take advantage of the great work done in the competition and submit a paper to the DCASE Workshop, to be held in New York in October (submission deadline July 5th and 12th). The organisers will be there and we’d be pleased to meet you!
If you are participating to this competition or using the FSDKaggle2019 dataset or baseline code, please cite the following paper:
> Eduardo Fonseca, Manoj Plakal, Frederic Font, Daniel P. W. Ellis, and Xavier Serra. \"Audio tagging with noisy labels and minimal supervision\". Submitted to DCASE2019 Workshop, 2019. URL: https://arxiv.org/abs/1906.02975
\n\n
\nBest of luck, on behalf of the organizers! We'll post another message when the private LB is out.
Eduardo Fonseca, MTG-UPF, Barcelona https://twitter.com/edfonseca_
Manoj Plakal, Google's Sound Understanding, New York
Frederic Font, MTG-UPF, Barcelona
Dan Ellis, Google's Sound Understanding, New York
Dear participants,
\n\nThe competition has concluded, and Stage 2 has begun. Kaggle is re-running kernels on the hidden test set. The extended deadline is to let you know that the competition is still active, and the leaderboard is still being updated. You can expect final results in a couple of weeks.
\n\nWe’d like to thank everyone for your passionate participation. It has been a few intense months where both participants and organizers have learnt a lot. 881 teams and 12,869 submissions! Way more than last year’s edition of the competition. We would like to give special thanks to those who contributed in the Discussion forums sharing their knowledge, and also to those who submitted through the DCASE system. This will have an impact on the research community.
\n\nSome announcements:
\n\nIn the coming days, submissions and code may be reviewed for the winners and for some other teams.
You can update your technical reports to the DCASE Challenge until 17 June 2019 (AoE). http://dcase.community/challenge2019/submission
After all the technical reports have been updated, the organizers will begin the review of the Judges’ Award submissions.
Regardless whether you submitted a technical report or not, we encourage you all to take advantage of the great work done in the competition and submit a paper to the DCASE Workshop, to be held in New York in October (submission deadline July 5th and 12th). The organisers will be there and we’d be pleased to meet you!
If you are participating to this competition or using the FSDKaggle2019 dataset or baseline code, please cite the following paper:
> Eduardo Fonseca, Manoj Plakal, Frederic Font, Daniel P. W. Ellis, and Xavier Serra. \"Audio tagging with noisy labels and minimal supervision\". Submitted to DCASE2019 Workshop, 2019. URL: https://arxiv.org/abs/1906.02975
\n\n
\nBest of luck, on behalf of the organizers! We'll post another message when the private LB is out.
Eduardo Fonseca, MTG-UPF, Barcelona https://twitter.com/edfonseca_
Manoj Plakal, Google's Sound Understanding, New York
Frederic Font, MTG-UPF, Barcelona
Dan Ellis, Google's Sound Understanding, New York
As first time kaggler, I'm not sure about the second stage. Do you mean there is nothing else for kagglers to do in the second stage? Kaggle will run on the new test set as the private LB score automatically?
", "rawMarkdown": "As first time kaggler, I'm not sure about the second stage. Do you mean there is nothing else for kagglers to do in the second stage? Kaggle will run on the new test set as the private LB score automatically?", "votes": null }, { "id": "554576", "postDate": "06/17/2019 17:22:29", "content": "Yes, as Eduardo mentioned in his message, Kaggle is re-running kernels on the hidden test set. We will also be reviewing the submissions of several teams. You do not need to do anything more with your submissions.
", "rawMarkdown": "Yes, as Eduardo mentioned in his message, Kaggle is re-running kernels on the hidden test set. We will also be reviewing the submissions of several teams. You do not need to do anything more with your submissions.", "votes": null } ], "comments": [ { "id": 554486, "author_name": "smallfade", "author_url": "", "post_date": "06/17/2019 13:39:23", "content": "As first time kaggler, I'm not sure about the second stage. Do you mean there is nothing else for kagglers to do in the second stage? Kaggle will run on the new test set as the private LB score automatically?
", "votes": null, "replies": [ { "id": 554576, "author_name": "plakal", "author_url": "", "post_date": "06/17/2019 17:22:29", "content": "Yes, as Eduardo mentioned in his message, Kaggle is re-running kernels on the hidden test set. We will also be reviewing the submissions of several teams. You do not need to do anything more with your submissions.
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "552153": "Dear participants,\n\nThe competition has concluded, and Stage 2 has begun. Kaggle is re-running kernels on the hidden test set. The extended deadline is to let you know that the competition is still active, and the leaderboard is still being updated. You can expect final results in a couple of weeks.\n\nWe’d like to thank everyone for your passionate participation. It has been a few intense months where both participants and organizers have learnt a lot. 881 teams and 12,869 submissions! Way more than last year’s edition of the competition. We would like to give **special thanks** to those who contributed in the Discussion forums sharing their knowledge, and also to those who submitted through the DCASE system. This will have an impact on the research community.\n\nSome announcements:\n\n- In the coming days, submissions and code may be reviewed for the winners and for some other teams.\n\n- You can update your technical reports to the DCASE Challenge until 17 June 2019 (AoE). [http://dcase.community/challenge2019/submission](http://dcase.community/challenge2019/submission)\n\n- After all the technical reports have been updated, the organizers will begin the review of the [Judges’ Award](https://www.kaggle.com/c/freesound-audio-tagging-2019/overview/judges-award) submissions.\n\n- Regardless whether you submitted a technical report or not, we encourage you all to take advantage of the great work done in the competition and **submit a paper to the [DCASE Workshop](http://dcase.community/workshop2019/)**, to be held in New York in October (submission deadline July 5th and 12th). The organisers will be there and we’d be pleased to meet you!\n\n- If you are participating to this competition or using the FSDKaggle2019 dataset or baseline code, please cite the following paper:\n\n> Eduardo Fonseca, Manoj Plakal, Frederic Font, Daniel P. W. Ellis, and Xavier Serra. \"Audio tagging with noisy labels and minimal supervision\". Submitted to DCASE2019 Workshop, 2019. URL: [https://arxiv.org/abs/1906.02975](https://arxiv.org/abs/1906.02975)\n\n