When submit my submission.csv to Competition, \nan error occurs(No additional details provided for this error). \nAnd the Status shows Submission CSV Not Found.
\n\nHow can I solve such problem?\nThanks a lot!
", "messages": [ { "id": 564833, "postDate": "2019-06-30T04:17:00.833Z", "content": "Make sure you have index=False in your df.to_csv('submission.csv', index=False).
I'm sure, but still has the same problem.
", "rawMarkdown": "I'm sure, but still has the same problem. " }, { "id": 564898, "postDate": "2019-06-30T06:21:11.800Z", "rawMarkdown": "", "votes": 1, "isDeleted": true }, { "id": 564902, "postDate": "2019-06-30T06:26:43.063Z", "content": "Thank you bro. \nNow I try to submit like https://www.kaggle.com/abhishek/pytorch-inference-kernel.
", "rawMarkdown": "Thank you bro. \nNow I try to submit like https://www.kaggle.com/abhishek/pytorch-inference-kernel." }, { "id": 564929, "postDate": "2019-06-30T07:15:07.593Z", "content": "Then, how did you solve that?
", "rawMarkdown": "Then, how did you solve that? " }, { "id": 564943, "postDate": "2019-06-30T07:44:17.943Z", "content": "I loaded all processed images to a dictionary (I want to use it for many models), that leads to out of memory when running on private test set (successful on public test set). I think so
", "rawMarkdown": "I loaded all processed images to a dictionary (I want to use it for many models), that leads to out of memory when running on private test set (successful on public test set). I think so", "votes": 5 }, { "id": 571935, "postDate": "2019-07-10T08:46:39.867Z", "content": "@nguyenbadung I know that feeling too :smile:\nI actually started from Abhishek's kernels. I'm training my model also on past competition data but whenever I try to submit... \"CSV not Found\". I tried to change batch size and/or image size within the inference kernel. No luck.\nThe problem was that I needed to change my training phase 😅 So I decreased image and batch size in that kernel and finally, it worked! (After ~15 wasted submissions)\nHas anyone managed to use a proper image size when using past comp data?
When submit my submission.csv to Competition, \nan error occurs(No additional details provided for this error). \nAnd the Status shows Submission CSV Not Found.
\n\nHow can I solve such problem?\nThanks a lot!
", "rawMarkdown": "When submit my submission.csv to Competition, \nan error occurs(No additional details provided for this error). \nAnd the Status shows Submission CSV Not Found.\n\nHow can I solve such problem?\nThanks a lot!", "votes": 2 }, { "id": 571892, "postDate": "2019-07-10T07:50:52.807Z", "content": "how do we resolve this error?
", "rawMarkdown": "how do we resolve this error? " }, { "id": 566639, "postDate": "2019-07-02T12:08:36.127Z", "content": "Facing another error submission take longer time but cannot get public scores
", "rawMarkdown": "Facing another error submission take longer time but cannot get public scores ", "replies": [ { "id": 566661, "postDate": "2019-07-02T12:33:20.383Z", "content": "Wait for a moment. My submission run about 1.5 hours with 10 tta.
", "rawMarkdown": "Wait for a moment. My submission run about 1.5 hours with 10 tta." } ] }, { "id": 566266, "postDate": "2019-07-02T01:31:32.903Z", "content": "Hi~, if I cannot train my model on PC, is it means that I cannot submit any work of mine ?\nI trained my model in kernel and I have tried almost any way to submit online, but it didn`t work..\nThe solution that you mentioned , it means we must train locally and submit it online, is it right?\nThe competition bother me a lot ...hoping for your reply~
", "rawMarkdown": "Hi~, if I cannot train my model on PC, is it means that I cannot submit any work of mine ?\nI trained my model in kernel and I have tried almost any way to submit online, but it didn`t work..\nThe solution that you mentioned , it means we must train locally and submit it online, is it right?\nThe competition bother me a lot ...hoping for your reply~" }, { "id": 564331, "postDate": "2019-06-29T09:34:56.497Z", "content": "Same problem trying to submit a offline generated csv file and trying to load it creating a dataset, loading it and saving it as a submission.
", "rawMarkdown": "Same problem trying to submit a offline generated csv file and trying to load it creating a dataset, loading it and saving it as a submission.", "replies": [ { "id": 564335, "postDate": "2019-06-29T09:38:38.173Z", "content": "You cannot submit offline generated CSV file as submission in this competition as your model needs to run on stage-2 dataset.
", "rawMarkdown": "You cannot submit offline generated CSV file as submission in this competition as your model needs to run on stage-2 dataset.", "votes": 1 }, { "id": 564337, "postDate": "2019-06-29T09:39:47.800Z", "content": "see this kernel on how to do it properly: https://www.kaggle.com/abhishek/pytorch-inference-kernel
", "rawMarkdown": "see this kernel on how to do it properly: https://www.kaggle.com/abhishek/pytorch-inference-kernel", "votes": 5 }, { "id": 565769, "postDate": "2019-07-01T11:07:33.950Z", "content": "Let me explain in more detail the submission procedure that fails:
\n\n1) I download the dataset\n2) I make some local preprocessing of the dataset\n3) I create a new dataset with the preprocessed images\n4) I create a kernel to treat such derived dataset
\n\nThe same kernel using the original dataset works properly.
\n\nCould someone explain to me what I am missing?
\n\nSorry for the annoyance!
", "rawMarkdown": "Let me explain in more detail the submission procedure that fails:\n\n1) I download the dataset\n2) I make some local preprocessing of the dataset\n3) I create a new dataset with the preprocessed images\n4) I create a kernel to treat such derived dataset\n\nThe same kernel using the original dataset works properly. \n\nCould someone explain to me what I am missing?\n\nSorry for the annoyance!", "votes": 1 } ] } ], "comments": [ { "id": 564833, "author_name": "Bryan Arnold", "author_url": "", "post_date": "2019-06-30T04:17:00.833000", "content": "Make sure you have index=False in your df.to_csv('submission.csv', index=False).
I'm sure, but still has the same problem.
", "votes": 0, "replies": [] }, { "id": 564898, "author_name": "", "author_url": "", "post_date": "2019-06-30T06:21:11.800000", "content": "", "votes": 1, "replies": [] }, { "id": 564902, "author_name": "", "author_url": "", "post_date": "2019-06-30T06:26:43.063000", "content": "Thank you bro. \nNow I try to submit like https://www.kaggle.com/abhishek/pytorch-inference-kernel.
", "votes": 0, "replies": [] }, { "id": 564929, "author_name": "Pooh", "author_url": "", "post_date": "2019-06-30T07:15:07.593000", "content": "Then, how did you solve that?
", "votes": 0, "replies": [] }, { "id": 564943, "author_name": "DungNB", "author_url": "", "post_date": "2019-06-30T07:44:17.943000", "content": "I loaded all processed images to a dictionary (I want to use it for many models), that leads to out of memory when running on private test set (successful on public test set). I think so
", "votes": 5, "replies": [] }, { "id": 571935, "author_name": "Federico Raimondi Cominesi", "author_url": "", "post_date": "2019-07-10T08:46:39.867000", "content": "@nguyenbadung I know that feeling too :smile:\nI actually started from Abhishek's kernels. I'm training my model also on past competition data but whenever I try to submit... \"CSV not Found\". I tried to change batch size and/or image size within the inference kernel. No luck.\nThe problem was that I needed to change my training phase 😅 So I decreased image and batch size in that kernel and finally, it worked! (After ~15 wasted submissions)\nHas anyone managed to use a proper image size when using past comp data?
how do we resolve this error?
", "votes": 0, "replies": [] }, { "id": 566639, "author_name": "Mohammad Ali Abbas", "author_url": "", "post_date": "2019-07-02T12:08:36.127000", "content": "Facing another error submission take longer time but cannot get public scores
", "votes": 0, "replies": [ { "id": 566661, "author_name": "", "author_url": "", "post_date": "2019-07-02T12:33:20.383000", "content": "Wait for a moment. My submission run about 1.5 hours with 10 tta.
", "votes": 0, "replies": [] } ] }, { "id": 566266, "author_name": "", "author_url": "", "post_date": "2019-07-02T01:31:32.903000", "content": "Hi~, if I cannot train my model on PC, is it means that I cannot submit any work of mine ?\nI trained my model in kernel and I have tried almost any way to submit online, but it didn`t work..\nThe solution that you mentioned , it means we must train locally and submit it online, is it right?\nThe competition bother me a lot ...hoping for your reply~
", "votes": 0, "replies": [] }, { "id": 564331, "author_name": "jordi.delatorre", "author_url": "", "post_date": "2019-06-29T09:34:56.497000", "content": "Same problem trying to submit a offline generated csv file and trying to load it creating a dataset, loading it and saving it as a submission.
", "votes": 0, "replies": [ { "id": 564335, "author_name": "Abhishek Thakur", "author_url": "", "post_date": "2019-06-29T09:38:38.173000", "content": "You cannot submit offline generated CSV file as submission in this competition as your model needs to run on stage-2 dataset.
", "votes": 1, "replies": [] }, { "id": 564337, "author_name": "Abhishek Thakur", "author_url": "", "post_date": "2019-06-29T09:39:47.800000", "content": "see this kernel on how to do it properly: https://www.kaggle.com/abhishek/pytorch-inference-kernel
", "votes": 5, "replies": [] }, { "id": 565769, "author_name": "jordi.delatorre", "author_url": "", "post_date": "2019-07-01T11:07:33.950000", "content": "Let me explain in more detail the submission procedure that fails:
\n\n1) I download the dataset\n2) I make some local preprocessing of the dataset\n3) I create a new dataset with the preprocessed images\n4) I create a kernel to treat such derived dataset
\n\nThe same kernel using the original dataset works properly.
\n\nCould someone explain to me what I am missing?
\n\nSorry for the annoyance!
", "votes": 1, "replies": [] } ] } ], "raw_markdown_by_id": { "564833": "Make sure you have `index=False` in your `df.to_csv('submission.csv', index=False)`.", "564179": "When submit my submission.csv to Competition, \nan error occurs(No additional details provided for this error). \nAnd the Status shows Submission CSV Not Found.\n\nHow can I solve such problem?\nThanks a lot!", "571892": "how do we resolve this error? ", "566639": "Facing another error submission take longer time but cannot get public scores ", "566266": "Hi~, if I cannot train my model on PC, is it means that I cannot submit any work of mine ?\nI trained my model in kernel and I have tried almost any way to submit online, but it didn`t work..\nThe solution that you mentioned , it means we must train locally and submit it online, is it right?\nThe competition bother me a lot ...hoping for your reply~", "564331": "Same problem trying to submit a offline generated csv file and trying to load it creating a dataset, loading it and saving it as a submission." } }