![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2698359%2F04d95f374d6670bb4114b4346c961ea3%2FCapture.PNG?generation=1587647954398817&alt=media\")
I still can't understand, how to submit in this competition .\nWhat I understand is that the test data are only available when I submit my kernel , but it's not working also .
What am I missing ?
", "messages": [ { "id": 837696, "postDate": "2020-05-08T01:15:24.003Z", "content": "Did you figure it out? I still cannot access the test data when I commit my notebook. I see that directory doesn't exist when I commit with this line:\n\n!ls ../input/prostate-cancer-grade-assessment/test_images/\nls: cannot access '../input/prostate-cancer-grade-assessment/test_images/': No such file or directory\n\nAnd, I'm still getting the truncated test.csv file.
Yea , It was a bug in my code , and it should be too in your , you can do what I did :\nbuild your model, save the model and use this kernel to submit
", "rawMarkdown": "Yea , It was a bug in my code , and it should be too in your , you can do what I did :\nbuild your model, save the model and use [this](https://www.kaggle.com/ateplyuk/panda-starter-infer) kernel to submit " }, { "id": 838745, "postDate": "2020-05-08T18:53:50.767Z", "content": "Oh, I see now. Reading other discussions finally made it clear Kaggle's submission process. Submitting a CSV actually runs the code again and this is where it can access the test data. So, output a dummy submission.csv (oh, and turn internet off) and submit that so Kaggle will run your notebook again. That wasn't obvious to me at first in case anyone else has issues.
", "rawMarkdown": "Oh, I see now. Reading other discussions finally made it clear Kaggle's submission process. Submitting a CSV actually runs the code again and this is where it can access the test data. So, output a dummy submission.csv (oh, and turn internet off) and submit that so Kaggle will run your notebook again. That wasn't obvious to me at first in case anyone else has issues.", "votes": 1 } ] }, { "id": 817893, "postDate": "2020-04-23T13:59:00.250Z", "content": "simply follow this kernel.
\n\nIn order to make sure everything is working correctly, first test on train data like this.
\n\n\nif os.path.exists('../input/prostate-cancer-grade-assessment/train_images')\n
If everything works properly, then while submitting replace train_images with test_images.
Yes it is working on the train, but not on the test ...
", "rawMarkdown": "Yes it is working on the train, but not on the test ..." }, { "id": 817958, "postDate": "2020-04-23T15:03:20.760Z", "content": "You might have done something wrong. Can't tell without looking at your kernel.
", "rawMarkdown": "You might have done something wrong. Can't tell without looking at your kernel." }, { "id": 817980, "postDate": "2020-04-23T15:16:07.650Z", "content": "It's just this code here\nas you can see train is working but test is not :()
", "rawMarkdown": "It's just this code [here](https://www.kaggle.com/yaheaal/test-file?scriptVersionId=32542217)\nas you can see train is working but test is not :()" }, { "id": 818017, "postDate": "2020-04-23T15:36:35.740Z", "content": "You are getting it wrong, test_images folder is hidden, you need to make submission in order to access it.
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3982638%2F99da7820cf7c7f9b2bdfd31c1114b62c%2FScreenshot%202020-04-23%20at%209.09.33%20PM.png?generation=1587656460513338&alt=media\")
There is no way you can access it by simple commiting the kernel.
", "rawMarkdown": "You are getting it wrong, `test_images` folder is hidden, you need to make submission in order to access it. \n\n\n\nThere is no way you can access it by simple commiting the kernel.", "votes": 1 }, { "id": 818031, "postDate": "2020-04-23T15:42:23.417Z", "content": "ahaaa I will try to do that .\nYes put it's written in pytorch , so because I know nothing about pytorch it is hard for me to understand it :(\nThanks for helping :)
", "rawMarkdown": "ahaaa I will try to do that .\nYes put it's written in pytorch , so because I know nothing about pytorch it is hard for me to understand it :(\nThanks for helping :)" }, { "id": 818036, "postDate": "2020-04-23T15:44:34.860Z", "content": "Welcome:)
", "rawMarkdown": "Welcome:)" } ] }, { "id": 817856, "postDate": "2020-04-23T13:20:14.403Z", "content": "I still can't understand, how to submit in this competition .\nWhat I understand is that the test data are only available when I submit my kernel , but it's not working also .
What am I missing ?
", "rawMarkdown": "I still can't understand, how to submit in this competition .\nWhat I understand is that the test data are only available when I submit my kernel , but it's not working also .\n\nWhat am I missing ?\n" } ], "comments": [ { "id": 837696, "author_name": "juno0812", "author_url": "", "post_date": "2020-05-08T01:15:24.003000", "content": "Did you figure it out? I still cannot access the test data when I commit my notebook. I see that directory doesn't exist when I commit with this line:\n\n!ls ../input/prostate-cancer-grade-assessment/test_images/\nls: cannot access '../input/prostate-cancer-grade-assessment/test_images/': No such file or directory\n\nAnd, I'm still getting the truncated test.csv file.
Yea , It was a bug in my code , and it should be too in your , you can do what I did :\nbuild your model, save the model and use this kernel to submit
", "votes": 0, "replies": [] }, { "id": 838745, "author_name": "juno0812", "author_url": "", "post_date": "2020-05-08T18:53:50.767000", "content": "Oh, I see now. Reading other discussions finally made it clear Kaggle's submission process. Submitting a CSV actually runs the code again and this is where it can access the test data. So, output a dummy submission.csv (oh, and turn internet off) and submit that so Kaggle will run your notebook again. That wasn't obvious to me at first in case anyone else has issues.
", "votes": 1, "replies": [] } ] }, { "id": 817893, "author_name": "Dracarys", "author_url": "", "post_date": "2020-04-23T13:59:00.250000", "content": "simply follow this kernel.
\n\nIn order to make sure everything is working correctly, first test on train data like this.
\n\n\nif os.path.exists('../input/prostate-cancer-grade-assessment/train_images')\n
If everything works properly, then while submitting replace train_images with test_images.
Yes it is working on the train, but not on the test ...
", "votes": 0, "replies": [] }, { "id": 817958, "author_name": "Dracarys", "author_url": "", "post_date": "2020-04-23T15:03:20.760000", "content": "You might have done something wrong. Can't tell without looking at your kernel.
", "votes": 0, "replies": [] }, { "id": 817980, "author_name": "Yaheaal", "author_url": "", "post_date": "2020-04-23T15:16:07.650000", "content": "It's just this code here\nas you can see train is working but test is not :()
", "votes": 0, "replies": [] }, { "id": 818017, "author_name": "Dracarys", "author_url": "", "post_date": "2020-04-23T15:36:35.740000", "content": "You are getting it wrong, test_images folder is hidden, you need to make submission in order to access it.
There is no way you can access it by simple commiting the kernel.
", "votes": 1, "replies": [] }, { "id": 818031, "author_name": "Yaheaal", "author_url": "", "post_date": "2020-04-23T15:42:23.417000", "content": "ahaaa I will try to do that .\nYes put it's written in pytorch , so because I know nothing about pytorch it is hard for me to understand it :(\nThanks for helping :)
", "votes": 0, "replies": [] }, { "id": 818036, "author_name": "Dracarys", "author_url": "", "post_date": "2020-04-23T15:44:34.860000", "content": "Welcome:)
", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "837696": "Did you figure it out? I still cannot access the test data when I commit my notebook. I see that directory doesn't exist when I commit with this line:\n```\n!ls ../input/prostate-cancer-grade-assessment/test_images/\nls: cannot access '../input/prostate-cancer-grade-assessment/test_images/': No such file or directory\n```\nAnd, I'm still getting the truncated `test.csv` file.", "817893": "simply follow [this](https://www.kaggle.com/rohitsingh9990/panda-resnext-inference) kernel.\n\nIn order to make sure everything is working correctly, first test on train data like this.\n\n```\nif os.path.exists('../input/prostate-cancer-grade-assessment/train_images')\n```\n\nIf everything works properly, then while submitting replace `train_images` with `test_images`.", "817856": "I still can't understand, how to submit in this competition .\nWhat I understand is that the test data are only available when I submit my kernel , but it's not working also .\n\nWhat am I missing ?\n" } }