Where communities thrive

Join over 1.5M+ people
Join over 100K+ communities
Free without limits
Create your own community

Explore more communities

crowdAI/spotify-sequential-skip-prediction-challenge

People

Repo info

Activity

brianbrost

@brianbrost

a7193e27165ab849fb8e70156d9aa265 training_set_5.tar.gz

brianbrost

@brianbrost

65d3b5731f1f735ccb8f7de1128c3354 training_set_6.tar.gz

b2e6e6c0989b9995672cc92219ac4bd8 training_set_7.tar.gz

1d716c77bcc64ca197372a89c7963d3d training_set_8.tar.gz

58f1e0b1e3d2c91edef3903199f15e9a training_set_9.tar.gz

@rstudent_gitlab please let me know if your checksums are different, and if you can extract any of the split files?

RStudent

@rstudent_gitlab

I know i am being a bother. Could someone else please who has downloaded the file successfully confirm the md5sum please. At least i will know the corruption error is valid

brianbrost

@brianbrost

I just downloaded those from the competition website, so the md5sums are the ones I would expect anyone else to get too. Are some of your checksums different, or are all of them different?

@rstudent_gitlab

RStudent

@rstudent_gitlab

Thank you so much for helping me out on this @brianbrost , @spMohanty . Well my md5sums are different for the main and the splits. For example for ~/workspace/music/train$ md5sum training_set_1.tar.gz
9bdea6a8c4a9e47b47bace227bad252f training_set_1.tar.gz

md5sum training_set_9.tar.gz
b1b4393806a9d90711e619bfd08dd2f7 training_set_9.tar.gz

So far i have tried , on click download, wget, CurL, no luck so far. Could you please give me actual size of these files. I will try clean downloads again and will monitor them. Maybe this time I will get them to work :) Thanks a lot for looking into it

brianbrost

@brianbrost

@rstudent_gitlab

with ls -l, I get the following sizes:

6044161773 Dec 12 15:09 training_set_0.tar.gz
6044989394 Dec 12 15:15 training_set_1.tar.gz
6042349689 Dec 12 15:21 training_set_2.tar.gz
6043674073 Dec 12 15:27 training_set_3.tar.gz
6042105510 Dec 12 15:34 training_set_4.tar.gz
6043173901 Dec 12 15:40 training_set_5.tar.gz
6042906018 Dec 12 15:46 training_set_6.tar.gz
6046086003 Dec 12 15:52 training_set_7.tar.gz
6043512819 Dec 12 15:58 training_set_8.tar.gz
6045573656 Dec 12 16:04 training_set_9.tar.gz
60438525791 Dec 23 00:14 20181113_training_set.tar.gz

Hope you manage to figure out the problem, and let me know if there's any other way I can help. Unfortunately I don't have any useful suggestions, except to try to re-download, which I guess is what you're trying right now. Please let us know if you sort out the problem!

RStudent

@rstudent_gitlab

@brianbrost @spMohanty i cannot thank you enough. back in business using the wonder of modern technology: Aria2 https://aria2.github.io/

Now time to use the other wonder of modern technology: deep net. Thanks again

Joey

@joychengzhaoyue_twitter

I’m wondering the deadline is Jan 4th is in which time zone? Thanks!

brianbrost

@brianbrost

@spMohanty What timezone is the competition deadline currently set for?

SP Mohanty

@spMohanty

@brianbrost : Its in UTC. Looking at the deadline its on Jan 4th, 12:00 UTC

Sainath Adapa

@sainathadapa

Just want to get a confirmation: The objective is to predict if a track was played briefly (skip_2 being true). It is not to predict if the track was skipped (that is not_skipped being false). Am I correct?

brianbrost

@brianbrost

@sainathadapa that's correct. Apologies for the delay in replying. I don't get a notification if you don't use @ my username.

Sainath Adapa

@sainathadapa

thanks

andres ferraro

@andrebola__twitter

@brianbrost @spMohanty I get the following error when submitting a new solution: Error : An error occurred (404) when calling the HeadObject operation: Not Found

the previous submission got the same error so I think there might be some problem with the service

SP Mohanty

@spMohanty

Dear @/all ,
Some of the updates from our end did break a few things with the evaluator.
I am trying to get those fixed now, and will respond back in an hour or so.

SP Mohanty

@spMohanty

Okay, submissions are working again ! Sorry for the delay in the fix !

Best of luck !

Joey

@joychengzhaoyue_twitter

@spMohanty @brianbrost Hi, I'm wondering what's the portion of leaderboard public test in the entire test set? Thanks a lot!

SP Mohanty

@spMohanty

@joychengzhaoyue_twitter : The current leaderboard uses only 50% of the whole test set. The final leaderboard will either use the other 50% of the test set, or the whole test set. This is to be determined after a discussion with @brianbrost . But we have both computed anyway for all the submissions.

Joey

@joychengzhaoyue_twitter

Cool, Thanks a lot for the reply!

olivierjeunen

@olivierjeunen

@spMohanty @brianbrost
Hi, I have two small questions.
First: when will the leaderboard be final (showing scores including the other 50% of the test set)?
Second: the paper deadline is Friday January 11th 23:59 AoE, but when can we expect the paper reviews along with the notification?

Thanks already for your time!

brianbrost

@brianbrost

Hi, expect to have the paper notifications ready within 7 days, but we're meeting Monday and Tuesday to confirm the notification deadline

@olivierjeunen Also, the leaderboard will hopefully be finalized tomorrow, but this is also pending discussions tomorrow.

olivierjeunen

@olivierjeunen

Perfect, thank you for the clarification.

brianbrost

@brianbrost

@olivierjeunen The results on the private leaderboard have now been published. Congratulations on your placement! We actually expect to have the paper reviews and notifications completed within 4 days of the submission deadline (i.e. end of day the following Tuesday)

olivierjeunen

@olivierjeunen

@brianbrost, great. Thanks!
Just to check, will post-deadline submissions still be evaluated on the test set as well?

brianbrost

@brianbrost

@olivierjeunen I don't know if crowdAI will still evaluate submissions afterwards, but they won't count towards the final competition leaderboard position.

@spMohanty Can you answer this question?

Sainath Adapa

@sainathadapa

@brianbrost I'm using the following latex template for the paper \documentclass[sigconf]{acmart}. Can you tell me if this is the right one?

brianbrost

@brianbrost

@sainathadapa yeah that's correct!

Sainath Adapa

@sainathadapa

thanks

Sainath Adapa

@sainathadapa

@brianbrost Sorry about these questions at last moment, but can you confirm if the baseline model submitted by you that scored 0.537 MAA / 0.742 FPA is the following: the skipping behavior of the last track in the first half of the session is used as the prediction for all the tracks in the second half

olivierjeunen

@olivierjeunen

Dear @brianbrost , do you have any update regarding the workshop review and notification process?

brianbrost

@brianbrost

@olivierjeunen Notifications will be sent out today!

@sainathadapa Yes I can confirm that what you wrote for the baseline model is correct.

zwarshavsky

@zwarshavsky

Hello folks! I am utilizing the full track_features csv listing 1.8 million songs and its features, however I am unable to utilize the track id to run a spotify API query. The Track ID format should adhere to the following e.g. 6rqhFgbbKwnb9MLmUQDhG6, however the format I am seeing on the csv is e.g. t_cf0164dd-1531-4399-bfa6-dec19cd1fedc. Can anyone assist? Thank you!

brianbrost

@brianbrost

Hi @zwarshavsky, unfortunately for licensing reasons we we had to remove the public track IDs for the dataset. For this reason the IDs are internal to the dataset and can't be used in conjunction with the Spotify API. Sorry for the inconvenience!

Bruno Marinho

@BruunoMM

Hello guys! I'm working on a project in college about this challenge. I know it's late but it's more of a research / study. I tried downloading the dataset from CrowdAI but I can't login anymore so I don't have access to it :(

Could anyone help me get access to that dataset? Even if it's a small portion of the dataset, just so I can proceed with the research. I would really appreciate the help.

brianbrost

@brianbrost

Hi Bruno, crowdai has been replaced by a new entity called AI Crowd. The dataset should be available from

https://www.aicrowd.com/challenges/spotify-sequential-skip-prediction-challenge

Please let me know if you have any issues downloading the dataset from there