![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3234224%2F2c09c042a4bf4420dc517ac4cab7e7ed%2Fopening.PNG?generation=1562265266519049&alt=media\")
Hello, I am facing problem to submit kernel.
\n\nI tryed to:\n- Predict by batch\n- Reduce image size to 64x64\n- Reduce Keras epochs to 1..\n- del unused variables\n- Check the format of submission.csv with pandas\nBut the error still occuring.
\n\nHere is the predicting part of my code.
\n\nPATH = '../input/'
\n\n
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3234224%2Feb5d4f263a4bed16e9ef4332c59ade02%2Fbatch.png?generation=1562265384825880&alt=media\")
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3234224%2F39c27d1967fae1cb842669b95488d220%2Fsub.PNG?generation=1562265406466344&alt=media\")
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3234224%2F3f9025d5ebcc7fdef9097310be9b6df7%2Fsave_sub.PNG?generation=1562265420005283&alt=media\")
Thank you in advance for your help :)
", "messages": [ { "id": "568346", "postDate": "07/04/2019 18:37:25", "content": "Hello, I am facing problem to submit kernel.
\n\nI tryed to:\n- Predict by batch\n- Reduce image size to 64x64\n- Reduce Keras epochs to 1..\n- del unused variables\n- Check the format of submission.csv with pandas\nBut the error still occuring.
\n\nHere is the predicting part of my code.
\n\nPATH = '../input/'
\n\n
Thank you in advance for your help :)
", "rawMarkdown": "Hello, I am facing problem to submit kernel. \n\nI tryed to:\n- Predict by batch\n- Reduce image size to 64x64\n- Reduce Keras epochs to 1..\n- del unused variables\n- Check the format of submission.csv with pandas\nBut the error still occuring. \n\nHere is the predicting part of my code.\n\nPATH = '../input/'\n\n\n\n\n\n\n\n\n\nThank you in advance for your help :)", "votes": null }, { "id": "568800", "postDate": "07/05/2019 13:07:33", "content": "Where is it failing? Can you post the error message?
", "rawMarkdown": "Where is it failing? Can you post the error message?", "votes": null }, { "id": "568861", "postDate": "07/05/2019 14:53:08", "content": "Deleting a variable never reclaims memory. It only deletes a reference.\nTry doing gc.collect() calls between for loop.
", "rawMarkdown": "Deleting a variable never reclaims memory. It only deletes a reference.\nTry doing gc.collect() calls between for loop.", "votes": null }, { "id": "568870", "postDate": "07/05/2019 15:23:45", "content": "I think I have the same issue. The kernel gives no error on the public test but then, when I submit it exceeds the time limit.\nNotice that on public set it takes 500 seconds, so it should be fine :)\nI tried to recreate the kernel with no luck.\nNow I'm trying to submit with python script and not a notebook.
\n\nThis is really a waste of time...
", "rawMarkdown": "I think I have the same issue. The kernel gives no error on the public test but then, when I submit it exceeds the time limit.\nNotice that on public set it takes 500 seconds, so it should be fine :)\nI tried to recreate the kernel with no luck.\nNow I'm trying to submit with python script and not a notebook.\n\nThis is really a waste of time...", "votes": null }, { "id": "568882", "postDate": "07/05/2019 15:56:56", "content": "@Murlikrishnan Tripathi I forgot to add gc.collect(), but it did not imrpove that much memory usage. Still, I am not sure that the memory is an issue since it take about 7Gb on the submit kernel, I'm just saying I tryed to improved but did not change the error.\n@JT There is no error message else than \"No additional details ...\" when mouse is hover the red error logo.
\n\nIf anyone succeded to find a solution ping me :)
", "rawMarkdown": "Murlikrishnan Tripathi I forgot to add gc.collect(), but it did not imrpove that much memory usage. Still, I am not sure that the memory is an issue since it take about 7Gb on the submit kernel, I'm just saying I tryed to improved but did not change the error.\n@JT There is no error message else than \"No additional details ...\" when mouse is hover the red error logo.\n\nIf anyone succeded to find a solution ping me :)", "votes": null }, { "id": "568883", "postDate": "07/05/2019 15:58:40", "content": "Furethemore, our submission takes about 3 hours. Therefore, the 9h limit doesn't seems to be an issue.
", "rawMarkdown": "Furethemore, our submission takes about 3 hours. Therefore, the 9h limit doesn't seems to be an issue.", "votes": null }, { "id": "569827", "postDate": "07/07/2019 12:12:31", "content": "Hello everyone, after few days of research, I finaly found the problem. It was not about memory, running time or whatever. The problem was nothing else than my preprocessing function. It was croping black border to reduce the noise and maximise the signal. The preprocessing function was working perfectly on public test set and training set. However, I truely believe that the private test set got fully dark / black images. Resulting in a full croping (image[0:0] or something like that). Therefore, the predict fail and to_csv also.
\n\nHope it will help ! GL
\n\nVincent
", "rawMarkdown": "Hello everyone, after few days of research, I finaly found the problem. It was not about memory, running time or whatever. The problem was nothing else than my preprocessing function. It was croping black border to reduce the noise and maximise the signal. The preprocessing function was working perfectly on public test set and training set. However, I truely believe that the private test set got fully dark / black images. Resulting in a full croping (image[0:0] or something like that). Therefore, the predict fail and to_csv also.\n\nHope it will help ! GL\n\nVincent", "votes": null }, { "id": "569855", "postDate": "07/07/2019 12:58:09", "content": "I've been having similar issues when using some pre-processing .. runs fine for public test data but hidden test must contain something odd somewhere. Second submission fails with no description of the error :/ will have a closer look based in what you said
", "rawMarkdown": "I've been having similar issues when using some pre-processing .. runs fine for public test data but hidden test must contain something odd somewhere. Second submission fails with no description of the error :/ will have a closer look based in what you said", "votes": null }, { "id": "570481", "postDate": "07/08/2019 11:02:13", "content": "It seems that it was also mine problem. Thanks for the update!
", "rawMarkdown": "It seems that it was also mine problem. Thanks for the update!", "votes": null }, { "id": "570526", "postDate": "07/08/2019 12:08:40", "content": "I had the same problem with cropping functions! Removing that preprocessing solved the issue.\nRight now, I have another one and it says \"CSV not found\"...\nThis competition is more debugging the submission phase rather than learning
", "rawMarkdown": "I had the same problem with cropping functions! Removing that preprocessing solved the issue.\nRight now, I have another one and it says \"CSV not found\"...\nThis competition is more debugging the submission phase rather than learning", "votes": null }, { "id": "570774", "postDate": "07/08/2019 18:50:11", "content": "Hello everyone \nI also face the same problem when I submit submission.csv file. It shows status kernel Running for several hours, in the last says Error i (No additional details provided for this error ).\nwhereas all training and committing time is less than 6 hours. \nwhat is the solution to this problem??
", "rawMarkdown": "Hello everyone \nI also face the same problem when I submit submission.csv file. It shows status kernel Running for several hours, in the last says Error i (No additional details provided for this error ).\nwhereas all training and committing time is less than 6 hours. \nwhat is the solution to this problem??", "votes": null }, { "id": "570788", "postDate": "07/08/2019 19:01:24", "content": "Could it be that a larger private test set for scoring vs commit is causing it to run out of memory?
", "rawMarkdown": "Could it be that a larger private test set for scoring vs commit is causing it to run out of memory?", "votes": null }, { "id": "571319", "postDate": "07/09/2019 13:47:38", "content": "I also have the same problem with cropping function (However, the error message is “CSV Not Found”) ... If anybody use the cropping function in my public kernel, I can confirm that in the current kernel version, the function works fine as I was able to successfully have an LB score with the new function.
", "rawMarkdown": "I also have the same problem with cropping function (However, the error message is “CSV Not Found”) ... If anybody use the cropping function in my public kernel, I can confirm that in the current kernel version, the function works fine as I was able to successfully have an LB score with the new function.", "votes": null }, { "id": "572509", "postDate": "07/11/2019 02:38:19", "content": "Oh, I use the crop function in your kernel and also have your problem ’Submission CSV Not Found‘\nHave you solved it?
", "rawMarkdown": "Oh, I use the crop function in your kernel and also have your problem ’Submission CSV Not Found‘\nHave you solved it?", "votes": null }, { "id": "572512", "postDate": "07/11/2019 02:42:18", "content": "Reading other topics say that csv not found can also refer to the same issue. However I never get your error. GL
", "rawMarkdown": "Reading other topics say that csv not found can also refer to the same issue. However I never get your error. GL", "votes": null }, { "id": "585167", "postDate": "07/27/2019 03:55:09", "content": "I have also used the same cropping function, but getting the error: 'Submission CSV Not Found‘. Any ideas about this? @ratthachat
", "rawMarkdown": "I have also used the same cropping function, but getting the error: 'Submission CSV Not Found‘. Any ideas about this? @ratthachat", "votes": null }, { "id": "585174", "postDate": "07/27/2019 04:28:25", "content": "Hi @oromis Abhishek, did you use the newest cropping version in my kernel ? That’s function surely has no problem (since I already made like 70 submissions with it)
", "rawMarkdown": "Hi @oromis Abhishek, did you use the newest cropping version in my kernel ? That’s function surely has no problem (since I already made like 70 submissions with it)", "votes": null }, { "id": "594319", "postDate": "08/07/2019 20:37:12", "content": "Any update on this. I am facing same problem.
", "rawMarkdown": "Any update on this. I am facing same problem.", "votes": null }, { "id": "604763", "postDate": "08/21/2019 18:53:13", "content": "Yeah, I found that if I crop the retina images based on some radius estimation algorithm, the submission would fail. If I simply do a resize say to a square, things would work out just fine. So my issue was definitely with the data pre-processing part as well. Thanks for the tip! It worked for us.
", "rawMarkdown": "Yeah, I found that if I crop the retina images based on some radius estimation algorithm, the submission would fail. If I simply do a resize say to a square, things would work out just fine. So my issue was definitely with the data pre-processing part as well. Thanks for the tip! It worked for us.", "votes": null } ], "comments": [ { "id": 568800, "author_name": "monkeyorman", "author_url": "", "post_date": "07/05/2019 13:07:33", "content": "Where is it failing? Can you post the error message?
", "votes": null, "replies": [] }, { "id": 568861, "author_name": "mktripathi456", "author_url": "", "post_date": "07/05/2019 14:53:08", "content": "Deleting a variable never reclaims memory. It only deletes a reference.\nTry doing gc.collect() calls between for loop.
", "votes": null, "replies": [] }, { "id": 568870, "author_name": "raimonds1993", "author_url": "", "post_date": "07/05/2019 15:23:45", "content": "I think I have the same issue. The kernel gives no error on the public test but then, when I submit it exceeds the time limit.\nNotice that on public set it takes 500 seconds, so it should be fine :)\nI tried to recreate the kernel with no luck.\nNow I'm trying to submit with python script and not a notebook.
\n\nThis is really a waste of time...
", "votes": null, "replies": [] }, { "id": 568882, "author_name": "vincoux", "author_url": "", "post_date": "07/05/2019 15:56:56", "content": "@Murlikrishnan Tripathi I forgot to add gc.collect(), but it did not imrpove that much memory usage. Still, I am not sure that the memory is an issue since it take about 7Gb on the submit kernel, I'm just saying I tryed to improved but did not change the error.\n@JT There is no error message else than \"No additional details ...\" when mouse is hover the red error logo.
\n\nIf anyone succeded to find a solution ping me :)
", "votes": null, "replies": [] }, { "id": 568883, "author_name": "vincoux", "author_url": "", "post_date": "07/05/2019 15:58:40", "content": "Furethemore, our submission takes about 3 hours. Therefore, the 9h limit doesn't seems to be an issue.
", "votes": null, "replies": [] }, { "id": 569827, "author_name": "vincoux", "author_url": "", "post_date": "07/07/2019 12:12:31", "content": "Hello everyone, after few days of research, I finaly found the problem. It was not about memory, running time or whatever. The problem was nothing else than my preprocessing function. It was croping black border to reduce the noise and maximise the signal. The preprocessing function was working perfectly on public test set and training set. However, I truely believe that the private test set got fully dark / black images. Resulting in a full croping (image[0:0] or something like that). Therefore, the predict fail and to_csv also.
\n\nHope it will help ! GL
\n\nVincent
", "votes": null, "replies": [ { "id": 569855, "author_name": "taindow", "author_url": "", "post_date": "07/07/2019 12:58:09", "content": "I've been having similar issues when using some pre-processing .. runs fine for public test data but hidden test must contain something odd somewhere. Second submission fails with no description of the error :/ will have a closer look based in what you said
", "votes": null, "replies": [] }, { "id": 570481, "author_name": "rafzy15", "author_url": "", "post_date": "07/08/2019 11:02:13", "content": "It seems that it was also mine problem. Thanks for the update!
", "votes": null, "replies": [] }, { "id": 570526, "author_name": "raimonds1993", "author_url": "", "post_date": "07/08/2019 12:08:40", "content": "I had the same problem with cropping functions! Removing that preprocessing solved the issue.\nRight now, I have another one and it says \"CSV not found\"...\nThis competition is more debugging the submission phase rather than learning
", "votes": null, "replies": [] } ] }, { "id": 570774, "author_name": "yadavendra121", "author_url": "", "post_date": "07/08/2019 18:50:11", "content": "Hello everyone \nI also face the same problem when I submit submission.csv file. It shows status kernel Running for several hours, in the last says Error i (No additional details provided for this error ).\nwhereas all training and committing time is less than 6 hours. \nwhat is the solution to this problem??
", "votes": null, "replies": [] }, { "id": 570788, "author_name": "kbolaris", "author_url": "", "post_date": "07/08/2019 19:01:24", "content": "Could it be that a larger private test set for scoring vs commit is causing it to run out of memory?
", "votes": null, "replies": [] }, { "id": 571319, "author_name": "ratthachat", "author_url": "", "post_date": "07/09/2019 13:47:38", "content": "I also have the same problem with cropping function (However, the error message is “CSV Not Found”) ... If anybody use the cropping function in my public kernel, I can confirm that in the current kernel version, the function works fine as I was able to successfully have an LB score with the new function.
", "votes": null, "replies": [ { "id": 572509, "author_name": "icebergi", "author_url": "", "post_date": "07/11/2019 02:38:19", "content": "Oh, I use the crop function in your kernel and also have your problem ’Submission CSV Not Found‘\nHave you solved it?
", "votes": null, "replies": [] }, { "id": 572512, "author_name": "vincoux", "author_url": "", "post_date": "07/11/2019 02:42:18", "content": "Reading other topics say that csv not found can also refer to the same issue. However I never get your error. GL
", "votes": null, "replies": [] }, { "id": 585167, "author_name": "oromis", "author_url": "", "post_date": "07/27/2019 03:55:09", "content": "I have also used the same cropping function, but getting the error: 'Submission CSV Not Found‘. Any ideas about this? @ratthachat
", "votes": null, "replies": [] }, { "id": 585174, "author_name": "ratthachat", "author_url": "", "post_date": "07/27/2019 04:28:25", "content": "Hi @oromis Abhishek, did you use the newest cropping version in my kernel ? That’s function surely has no problem (since I already made like 70 submissions with it)
", "votes": null, "replies": [] } ] }, { "id": 594319, "author_name": "anndd3", "author_url": "", "post_date": "08/07/2019 20:37:12", "content": "Any update on this. I am facing same problem.
", "votes": null, "replies": [] }, { "id": 604763, "author_name": "kakirastern", "author_url": "", "post_date": "08/21/2019 18:53:13", "content": "Yeah, I found that if I crop the retina images based on some radius estimation algorithm, the submission would fail. If I simply do a resize say to a square, things would work out just fine. So my issue was definitely with the data pre-processing part as well. Thanks for the tip! It worked for us.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "568346": "Hello, I am facing problem to submit kernel. \n\nI tryed to:\n- Predict by batch\n- Reduce image size to 64x64\n- Reduce Keras epochs to 1..\n- del unused variables\n- Check the format of submission.csv with pandas\nBut the error still occuring. \n\nHere is the predicting part of my code.\n\nPATH = '../input/'\n\n\n\n\n\n\n\n\n\nThank you in advance for your help :)", "568800": "Where is it failing? Can you post the error message?", "568861": "Deleting a variable never reclaims memory. It only deletes a reference.\nTry doing gc.collect() calls between for loop.", "568870": "I think I have the same issue. The kernel gives no error on the public test but then, when I submit it exceeds the time limit.\nNotice that on public set it takes 500 seconds, so it should be fine :)\nI tried to recreate the kernel with no luck.\nNow I'm trying to submit with python script and not a notebook.\n\nThis is really a waste of time...", "568882": "Murlikrishnan Tripathi I forgot to add gc.collect(), but it did not imrpove that much memory usage. Still, I am not sure that the memory is an issue since it take about 7Gb on the submit kernel, I'm just saying I tryed to improved but did not change the error.\n@JT There is no error message else than \"No additional details ...\" when mouse is hover the red error logo.\n\nIf anyone succeded to find a solution ping me :)", "568883": "Furethemore, our submission takes about 3 hours. Therefore, the 9h limit doesn't seems to be an issue.", "569827": "Hello everyone, after few days of research, I finaly found the problem. It was not about memory, running time or whatever. The problem was nothing else than my preprocessing function. It was croping black border to reduce the noise and maximise the signal. The preprocessing function was working perfectly on public test set and training set. However, I truely believe that the private test set got fully dark / black images. Resulting in a full croping (image[0:0] or something like that). Therefore, the predict fail and to_csv also.\n\nHope it will help ! GL\n\nVincent", "569855": "I've been having similar issues when using some pre-processing .. runs fine for public test data but hidden test must contain something odd somewhere. Second submission fails with no description of the error :/ will have a closer look based in what you said", "570481": "It seems that it was also mine problem. Thanks for the update!", "570526": "I had the same problem with cropping functions! Removing that preprocessing solved the issue.\nRight now, I have another one and it says \"CSV not found\"...\nThis competition is more debugging the submission phase rather than learning", "570774": "Hello everyone \nI also face the same problem when I submit submission.csv file. It shows status kernel Running for several hours, in the last says Error i (No additional details provided for this error ).\nwhereas all training and committing time is less than 6 hours. \nwhat is the solution to this problem??", "570788": "Could it be that a larger private test set for scoring vs commit is causing it to run out of memory?", "571319": "I also have the same problem with cropping function (However, the error message is “CSV Not Found”) ... If anybody use the cropping function in my public kernel, I can confirm that in the current kernel version, the function works fine as I was able to successfully have an LB score with the new function.", "572509": "Oh, I use the crop function in your kernel and also have your problem ’Submission CSV Not Found‘\nHave you solved it?", "572512": "Reading other topics say that csv not found can also refer to the same issue. However I never get your error. GL", "585167": "I have also used the same cropping function, but getting the error: 'Submission CSV Not Found‘. Any ideas about this? @ratthachat", "585174": "Hi @oromis Abhishek, did you use the newest cropping version in my kernel ? That’s function surely has no problem (since I already made like 70 submissions with it)", "594319": "Any update on this. I am facing same problem.", "604763": "Yeah, I found that if I crop the retina images based on some radius estimation algorithm, the submission would fail. If I simply do a resize say to a square, things would work out just fine. So my issue was definitely with the data pre-processing part as well. Thanks for the tip! It worked for us." }, "source": "meta" }