{ "id": 109584, "title": "Submission CSV Not Found", "url": "/competitions/Kannada-MNIST/discussion/109584", "author_name": "", "post_date": "2019-09-20T11:44:31.044249300Z", "votes": 5, "comment_count": 17, "views": 0, "content": "

I keep encounter this error after I submit the submission csv file.. But the kernel seems ran normally and I can see the submission csv file is normal either. Did anyone has the same problem?

", "messages": [ { "id": "630558", "postDate": "09/20/2019 11:44:31", "content": "

I keep encounter this error after I submit the submission csv file.. But the kernel seems ran normally and I can see the submission csv file is normal either. Did anyone has the same problem?

", "rawMarkdown": "I keep encounter this error after I submit the submission csv file.. But the kernel seems ran normally and I can see the submission csv file is normal either. Did anyone has the same problem?", "votes": null }, { "id": "632842", "postDate": "09/24/2019 06:15:51", "content": "

I do. My kernel ran normally and I can see the submission.csv as well. However when submitting it I get the same error.

\n\n

Did you find any solution?

", "rawMarkdown": "I do. My kernel ran normally and I can see the submission.csv as well. However when submitting it I get the same error.\n\nDid you find any solution?", "votes": null }, { "id": "632858", "postDate": "09/24/2019 06:49:25", "content": "

After I downsize the model or reduce the batch size, the submission can be submitted normally. But it still takes a lot of time to pop out the score. Although I still don't know the actual root cause, but this method work fine by me.

", "rawMarkdown": "After I downsize the model or reduce the batch size, the submission can be submitted normally. But it still takes a lot of time to pop out the score. Although I still don't know the actual root cause, but this method work fine by me.", "votes": null }, { "id": "634024", "postDate": "09/25/2019 17:47:00", "content": "

I'm still having the same issue. https://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21093403

", "rawMarkdown": "I'm still having the same issue. https://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21093403", "votes": null }, { "id": "634139", "postDate": "09/25/2019 23:14:19", "content": "

Your model seems small enough.. This is weird. Maybe don't print file.read() in the end? I don't know whether this will help, but might worth for a shot.

", "rawMarkdown": "Your model seems small enough.. This is weird. Maybe don't print file.read() in the end? I don't know whether this will help, but might worth for a shot.", "votes": null }, { "id": "634147", "postDate": "09/25/2019 23:38:35", "content": "

Thanks for the suggestion. I tried removing the print and I still get the same error ☹️ \nhttps://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21102191

", "rawMarkdown": "Thanks for the suggestion. I tried removing the print and I still get the same error ☹️ \nhttps://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21102191", "votes": null }, { "id": "634150", "postDate": "09/25/2019 23:54:42", "content": "

Sorry to hear that :( This seems like something wrong with Kaggle.

", "rawMarkdown": "Sorry to hear that :( This seems like something wrong with Kaggle.", "votes": null }, { "id": "634253", "postDate": "09/26/2019 05:10:24", "content": "

It seems OK. I think I didn't change anything special (maybe I downsized my model a bit in the meantime) and it works for me now. It'd be great if any host or Kaggle team member could elaborate on that.

", "rawMarkdown": "It seems OK. I think I didn't change anything special (maybe I downsized my model a bit in the meantime) and it works for me now. It'd be great if any host or Kaggle team member could elaborate on that.", "votes": null }, { "id": "634306", "postDate": "09/26/2019 06:21:35", "content": "

Hi Chris, I think you created a “submission.csv” which contained 5,000 rows (hard-coded). When your kernel is submitted, your kernel would be run and checked against a different test set, which I think has more than 5,000 rows. So I think you need to tweak your codes related to submission to make sure they cater to different sizes of test sets. Hope this helps.

", "rawMarkdown": "Hi Chris, I think you created a “submission.csv” which contained 5,000 rows (hard-coded). When your kernel is submitted, your kernel would be run and checked against a different test set, which I think has more than 5,000 rows. So I think you need to tweak your codes related to submission to make sure they cater to different sizes of test sets. Hope this helps.", "votes": null }, { "id": "634382", "postDate": "09/26/2019 08:04:21", "content": "
\n

your kernel would be run and checked against a different test set

\n
\n\n

Why?

", "rawMarkdown": "> your kernel would be run and checked against a **different test set**\n\nWhy?", "votes": null }, { "id": "634441", "postDate": "09/26/2019 09:38:50", "content": "

This is mentioned in the competition overview as below:

\n\n

“This competition uses the same format as the MNIST competition in terms of how the data is structured, but it's different in that it is a synchronous re-run Kernels competition. You write your code in a Kaggle Notebook, and when you submit the results, your code is scored on both the public test set, as well as a private (unseen) test set.”

", "rawMarkdown": "This is mentioned in the competition overview as below:\n\n“This competition uses the same format as the MNIST competition in terms of how the data is structured, but it's different in that it is a synchronous re-run Kernels competition. You write your code in a Kaggle Notebook, and when you submit the results, your code is scored on both the public test set, as well as a private (unseen) test set.”", "votes": null }, { "id": "634444", "postDate": "09/26/2019 09:58:44", "content": "

Oh, I see, thanks @rhodiumbeng!

", "rawMarkdown": "Oh, I see, thanks @rhodiumbeng!", "votes": null }, { "id": "634452", "postDate": "09/26/2019 10:10:23", "content": "

That was it, @rhodiumbeng ! Thanks!\nhttps://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21122459

", "rawMarkdown": "That was it, @rhodiumbeng ! Thanks!\nhttps://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21122459", "votes": null }, { "id": "635134", "postDate": "09/27/2019 07:29:49", "content": "

I fixed this by rezising test dataset using \" np.resize((-1, 28, 28, 1) \". Before, I was using test dataset's length. Hope it helps!

", "rawMarkdown": "I fixed this by rezising test dataset using \" np.resize((-1, 28, 28, 1) \". Before, I was using test dataset's length. Hope it helps!", "votes": null }, { "id": "644254", "postDate": "10/08/2019 14:34:43", "content": "

I just commented a posting.\nchange from Notebook to Script. Run and Commit. (it should run in the cloud)\nafter few minutes the output folder should be there\nor you can copy my kernel and test it.

\n\n

let me know if this helped\nAdrian

", "rawMarkdown": "I just commented a posting.\nchange from Notebook to Script. Run and Commit. (it should run in the cloud)\nafter few minutes the output folder should be there\nor you can copy my kernel and test it.\n\nlet me know if this helped\nAdrian", "votes": null }, { "id": "644904", "postDate": "10/09/2019 14:02:46", "content": "

\"\"

\n\n

can anyone plz help me ...csv file is present but i am keep encountering this error

", "rawMarkdown": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3806394%2F8eccd9c7f49bde929e87bbc455d71214%2FCapture.PNG?generation=1570629702552029&alt=media)\n\ncan anyone plz help me ...csv file is present but i am keep encountering this error", "votes": null }, { "id": "657968", "postDate": "10/25/2019 14:55:56", "content": "

i think just using Kaggle's kernel 'add dataset' button can load these .csv files

", "rawMarkdown": "i think just using Kaggle's kernel 'add dataset' button can load these .csv files", "votes": null }, { "id": "658409", "postDate": "10/26/2019 01:09:31", "content": "

You can try os.listdir('../input/'), then you can see the files.

", "rawMarkdown": "You can try os.listdir('../input/'), then you can see the files.", "votes": null } ], "comments": [ { "id": 632842, "author_name": "mtszkw", "author_url": "", "post_date": "09/24/2019 06:15:51", "content": "

I do. My kernel ran normally and I can see the submission.csv as well. However when submitting it I get the same error.

\n\n

Did you find any solution?

", "votes": null, "replies": [ { "id": 632858, "author_name": "xiejialun", "author_url": "", "post_date": "09/24/2019 06:49:25", "content": "

After I downsize the model or reduce the batch size, the submission can be submitted normally. But it still takes a lot of time to pop out the score. Although I still don't know the actual root cause, but this method work fine by me.

", "votes": null, "replies": [] } ] }, { "id": 634024, "author_name": "cdonohue", "author_url": "", "post_date": "09/25/2019 17:47:00", "content": "

I'm still having the same issue. https://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21093403

", "votes": null, "replies": [ { "id": 634139, "author_name": "xiejialun", "author_url": "", "post_date": "09/25/2019 23:14:19", "content": "

Your model seems small enough.. This is weird. Maybe don't print file.read() in the end? I don't know whether this will help, but might worth for a shot.

", "votes": null, "replies": [] }, { "id": 634147, "author_name": "cdonohue", "author_url": "", "post_date": "09/25/2019 23:38:35", "content": "

Thanks for the suggestion. I tried removing the print and I still get the same error ☹️ \nhttps://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21102191

", "votes": null, "replies": [] }, { "id": 634150, "author_name": "xiejialun", "author_url": "", "post_date": "09/25/2019 23:54:42", "content": "

Sorry to hear that :( This seems like something wrong with Kaggle.

", "votes": null, "replies": [] }, { "id": 634253, "author_name": "mtszkw", "author_url": "", "post_date": "09/26/2019 05:10:24", "content": "

It seems OK. I think I didn't change anything special (maybe I downsized my model a bit in the meantime) and it works for me now. It'd be great if any host or Kaggle team member could elaborate on that.

", "votes": null, "replies": [] }, { "id": 634306, "author_name": "rhodiumbeng", "author_url": "", "post_date": "09/26/2019 06:21:35", "content": "

Hi Chris, I think you created a “submission.csv” which contained 5,000 rows (hard-coded). When your kernel is submitted, your kernel would be run and checked against a different test set, which I think has more than 5,000 rows. So I think you need to tweak your codes related to submission to make sure they cater to different sizes of test sets. Hope this helps.

", "votes": null, "replies": [] }, { "id": 634382, "author_name": "mtszkw", "author_url": "", "post_date": "09/26/2019 08:04:21", "content": "
\n

your kernel would be run and checked against a different test set

\n
\n\n

Why?

", "votes": null, "replies": [] }, { "id": 634441, "author_name": "rhodiumbeng", "author_url": "", "post_date": "09/26/2019 09:38:50", "content": "

This is mentioned in the competition overview as below:

\n\n

“This competition uses the same format as the MNIST competition in terms of how the data is structured, but it's different in that it is a synchronous re-run Kernels competition. You write your code in a Kaggle Notebook, and when you submit the results, your code is scored on both the public test set, as well as a private (unseen) test set.”

", "votes": null, "replies": [] }, { "id": 634444, "author_name": "mtszkw", "author_url": "", "post_date": "09/26/2019 09:58:44", "content": "

Oh, I see, thanks @rhodiumbeng!

", "votes": null, "replies": [] }, { "id": 634452, "author_name": "cdonohue", "author_url": "", "post_date": "09/26/2019 10:10:23", "content": "

That was it, @rhodiumbeng ! Thanks!\nhttps://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21122459

", "votes": null, "replies": [] } ] }, { "id": 635134, "author_name": "mesuat", "author_url": "", "post_date": "09/27/2019 07:29:49", "content": "

I fixed this by rezising test dataset using \" np.resize((-1, 28, 28, 1) \". Before, I was using test dataset's length. Hope it helps!

", "votes": null, "replies": [] }, { "id": 644254, "author_name": "zinovadr", "author_url": "", "post_date": "10/08/2019 14:34:43", "content": "

I just commented a posting.\nchange from Notebook to Script. Run and Commit. (it should run in the cloud)\nafter few minutes the output folder should be there\nor you can copy my kernel and test it.

\n\n

let me know if this helped\nAdrian

", "votes": null, "replies": [] }, { "id": 644904, "author_name": "echo31", "author_url": "", "post_date": "10/09/2019 14:02:46", "content": "

\"\"

\n\n

can anyone plz help me ...csv file is present but i am keep encountering this error

", "votes": null, "replies": [ { "id": 658409, "author_name": "xiejialun", "author_url": "", "post_date": "10/26/2019 01:09:31", "content": "

You can try os.listdir('../input/'), then you can see the files.

", "votes": null, "replies": [] } ] }, { "id": 657968, "author_name": "niefengxxx", "author_url": "", "post_date": "10/25/2019 14:55:56", "content": "

i think just using Kaggle's kernel 'add dataset' button can load these .csv files

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "630558": "I keep encounter this error after I submit the submission csv file.. But the kernel seems ran normally and I can see the submission csv file is normal either. Did anyone has the same problem?", "632842": "I do. My kernel ran normally and I can see the submission.csv as well. However when submitting it I get the same error.\n\nDid you find any solution?", "632858": "After I downsize the model or reduce the batch size, the submission can be submitted normally. But it still takes a lot of time to pop out the score. Although I still don't know the actual root cause, but this method work fine by me.", "634024": "I'm still having the same issue. https://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21093403", "634139": "Your model seems small enough.. This is weird. Maybe don't print file.read() in the end? I don't know whether this will help, but might worth for a shot.", "634147": "Thanks for the suggestion. I tried removing the print and I still get the same error ☹️ \nhttps://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21102191", "634150": "Sorry to hear that :( This seems like something wrong with Kaggle.", "634253": "It seems OK. I think I didn't change anything special (maybe I downsized my model a bit in the meantime) and it works for me now. It'd be great if any host or Kaggle team member could elaborate on that.", "634306": "Hi Chris, I think you created a “submission.csv” which contained 5,000 rows (hard-coded). When your kernel is submitted, your kernel would be run and checked against a different test set, which I think has more than 5,000 rows. So I think you need to tweak your codes related to submission to make sure they cater to different sizes of test sets. Hope this helps.", "634382": "> your kernel would be run and checked against a **different test set**\n\nWhy?", "634441": "This is mentioned in the competition overview as below:\n\n“This competition uses the same format as the MNIST competition in terms of how the data is structured, but it's different in that it is a synchronous re-run Kernels competition. You write your code in a Kaggle Notebook, and when you submit the results, your code is scored on both the public test set, as well as a private (unseen) test set.”", "634444": "Oh, I see, thanks @rhodiumbeng!", "634452": "That was it, @rhodiumbeng ! Thanks!\nhttps://www.kaggle.com/cdonohue/kannadamnist?scriptVersionId=21122459", "635134": "I fixed this by rezising test dataset using \" np.resize((-1, 28, 28, 1) \". Before, I was using test dataset's length. Hope it helps!", "644254": "I just commented a posting.\nchange from Notebook to Script. Run and Commit. (it should run in the cloud)\nafter few minutes the output folder should be there\nor you can copy my kernel and test it.\n\nlet me know if this helped\nAdrian", "644904": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3806394%2F8eccd9c7f49bde929e87bbc455d71214%2FCapture.PNG?generation=1570629702552029&alt=media)\n\ncan anyone plz help me ...csv file is present but i am keep encountering this error", "657968": "i think just using Kaggle's kernel 'add dataset' button can load these .csv files", "658409": "You can try os.listdir('../input/'), then you can see the files." }, "source": "meta" }