![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F991320%2F0f924496fe4aedad00f876ab50e97846%2FScreen%20Shot%202019-12-25%20at%2010.16.23%20AM.png?generation=1577287267695326&alt=media\")
Notebook is running normally\nbut ”Submit to Competition“ Error occurred\nNotebook Threw Exception or Notebook Timeout \nwhat's wrong???
", "messages": [ { "id": "702698", "postDate": "12/25/2019 03:02:32", "content": "Notebook is running normally\nbut ”Submit to Competition“ Error occurred\nNotebook Threw Exception or Notebook Timeout \nwhat's wrong???
", "rawMarkdown": "Notebook is running normally\nbut ”Submit to Competition“ Error occurred\nNotebook Threw Exception or Notebook Timeout \nwhat's wrong???", "votes": null }, { "id": "702918", "postDate": "12/25/2019 10:59:33", "content": "facing same error
", "rawMarkdown": "facing same error", "votes": null }, { "id": "703089", "postDate": "12/25/2019 15:26:15", "content": "Ok here are few things that can go wrong:
\n\n1) Make sure you notebook is not connected to internet. You can disable this on the right side of the notebook \n
2) If you create local folders or other files make sure to remove them at the end of the notebook (except submission.csv): Below is the example how it should look like: ![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F991320%2F7d42e8cb9e9cb935cff757182638e033%2FScreen%20Shot%202019-12-25%20at%2010.18.38%20AM.png?generation=1577287344712997&alt=media\")
3) Make sure you have GPU hours left. Everyone gets 30h per week. You can check it by going to your profile and clicking \"edit profile\" and look for Quotes. Below is the example how it looks:\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F991320%2Fd6566b3b099e1b69c79a49e5f8ba1545%2FScreen%20Shot%202019-12-25%20at%2010.23.19%20AM.png?generation=1577287442881532&alt=media\")
4) When you submit the test files get replaced by bigger dataset (e.g 200k images). You should make sure that your inference run in less then 9h. You can try to run prediction for 200000 images by replicating kaggle configuration 1 gpu and 2 cpu with 15gb ram.
\n\nGood luck =)
", "rawMarkdown": "Ok here are few things that can go wrong:\n\n\n1) Make sure you notebook is not connected to internet. You can disable this on the right side of the notebook \n\n\n2) If you create local folders or other files make sure to remove them at the end of the notebook (except submission.csv): Below is the example how it should look like: \n\n3) Make sure you have GPU hours left. Everyone gets 30h per week. You can check it by going to your profile and clicking \"edit profile\" and look for Quotes. Below is the example how it looks:\n\n\n4) When you submit the test files get replaced by bigger dataset (e.g 200k images). You should make sure that your inference run in less then 9h. You can try to run prediction for 200000 images by replicating kaggle configuration 1 gpu and 2 cpu with 15gb ram.\n\nGood luck =)", "votes": null }, { "id": "703308", "postDate": "12/26/2019 01:38:14", "content": "\n", "rawMarkdown": "> **DrHB wrote:**\n> \n> Ok here are few things that can go wrong:\n> \n> \n> 1) Make sure you notebook is not connected to internet. You can disable this on the right side of the notebook \n> \n> \n> 2) If you create local folders or other files make sure to remove them at the end of the notebook (except submission.csv): Below is the example how it should look like: \n> \n> 3) Make sure you have GPU hours left. Everyone gets 30h per week. You can check it by going to your profile and clicking \"edit profile\" and look for Quotes. Below is the example how it looks:\n> \n> \n> 4) When you submit the test files get replaced by bigger dataset (e.g 200k images). You should make sure that your inference run in less then 9h. You can try to run prediction for 200000 images by replicating kaggle configuration 1 gpu and 2 cpu with 15gb ram.\n> \n> Good luck =) \n> \n1.Thanks.\n2.There are no problems in the first three points, I have done it.\n3.Fourth, is there 200k test images?", "votes": null }, { "id": "703317", "postDate": "12/26/2019 01:58:29", "content": "DrHB wrote:
\n \nOk here are few things that can go wrong:
\n \n1) Make sure you notebook is not connected to internet. You can disable this on the right side of the notebook \n
\n \n2) If you create local folders or other files make sure to remove them at the end of the notebook (except submission.csv): Below is the example how it should look like:
\n \n3) Make sure you have GPU hours left. Everyone gets 30h per week. You can check it by going to your profile and clicking \"edit profile\" and look for Quotes. Below is the example how it looks:\n
\n \n4) When you submit the test files get replaced by bigger dataset (e.g 200k images). You should make sure that your inference run in less then 9h. You can try to run prediction for 200000 images by replicating kaggle configuration 1 gpu and 2 cpu with 15gb ram.
\n \nGood luck =)
\n \n1.Thanks.\n 2.There are no problems in the first three points, I have done it.\n 3.Fourth, is there 200k test images?
\n
3) Yes, as stated in the data description page, the number of test images are roughly the same as the number of train images.
", "rawMarkdown": "3) Yes, as stated in the data description page, the number of test images are roughly the same as the number of train images.", "votes": null }, { "id": "703337", "postDate": "12/26/2019 02:51:08", "content": "@drhabib DrHB\nBecause I use the pre training model, which requires three channels, np..repeat (data, 3), the code is as follows:\n df_test_img = pd.read_parquet('/kaggle/input/bengaliai-cv19/test_image_data_{}.parquet'.format(i),dtype=np.uint) \n df_test_img.set_index('image_id', inplace=True)
\n\nX_test = resize(df_test_img, need_progress_bar=False)/255.0\nX_test = X_test.values.reshape(-1, reHEIGHT,reWIDTH)\nX_test = np.repeat(X_test,3)\nX_test = np.array(X_test).reshape(-1, reHEIGHT,reWIDTH, N_CHANNELS)\nHowever, the following error occurred: 《Notebook Exceeded Allowed Compute》
", "rawMarkdown": "drhabib DrHB\nBecause I use the pre training model, which requires three channels, np..repeat (data, 3), the code is as follows:\n df_test_img = pd.read_parquet('/kaggle/input/bengaliai-cv19/test_image_data_{}.parquet'.format(i),dtype=np.uint) \n df_test_img.set_index('image_id', inplace=True)\n\n X_test = resize(df_test_img, need_progress_bar=False)/255.0\n X_test = X_test.values.reshape(-1, reHEIGHT,reWIDTH)\n X_test = np.repeat(X_test,3)\n X_test = np.array(X_test).reshape(-1, reHEIGHT,reWIDTH, N_CHANNELS)\nHowever, the following error occurred: 《Notebook Exceeded Allowed Compute》", "votes": null }, { "id": "703363", "postDate": "12/26/2019 03:37:55", "content": "@TahsinReasat\ncomponents = ['consonant_diacritic', 'grapheme_root', 'vowel_diacritic']\ntarget=[] # model predictions placeholder\nrow_id=[] # row_id place holder\nfor i in range(4):\n df_test_img = pd.read_parquet('/kaggle/input/bengaliai-cv19/test_image_data_{}.parquet'.format(i)) \n df_test_img.set_index('image_id', inplace=True)
\n\nX_test = resize(df_test_img, need_progress_bar=False)/255\n#X_test = X_test.values.reshape(-1, reHEIGHT,reWIDTH)\n#X_test = np.repeat(X_test,3)\nX_test = np.repeat(X_test.values,3).reshape(-1, reHEIGHT,reWIDTH, N_CHANNELS)\n\nfor pred in preds_dict:\n preds_dict[pred]=np.argmax(model_dict[pred].predict(X_test), axis=1)\n\nfor k,id in enumerate(df_test_img.index.values): \n for i,comp in enumerate(components):\n id_sample=id+'_'+comp\n row_id.append(id_sample)\n target.append(preds_dict[comp][k])\ndel df_test_img\ndel X_test\ngc.collect()\ndf_sample = pd.DataFrame(\n {\n 'row_id': row_id,\n 'target':target\n },\n columns = ['row_id','target'] \n)\ndf_sample.to_csv('submission.csv',index=False)
\n\nNotebook Exceeded Allowed Compute!!!
", "rawMarkdown": "TahsinReasat\ncomponents = ['consonant_diacritic', 'grapheme_root', 'vowel_diacritic']\ntarget=[] # model predictions placeholder\nrow_id=[] # row_id place holder\nfor i in range(4):\n df_test_img = pd.read_parquet('/kaggle/input/bengaliai-cv19/test_image_data_{}.parquet'.format(i)) \n df_test_img.set_index('image_id', inplace=True)\n\n X_test = resize(df_test_img, need_progress_bar=False)/255\n #X_test = X_test.values.reshape(-1, reHEIGHT,reWIDTH)\n #X_test = np.repeat(X_test,3)\n X_test = np.repeat(X_test.values,3).reshape(-1, reHEIGHT,reWIDTH, N_CHANNELS)\n\n for pred in preds_dict:\n preds_dict[pred]=np.argmax(model_dict[pred].predict(X_test), axis=1)\n\n for k,id in enumerate(df_test_img.index.values): \n for i,comp in enumerate(components):\n id_sample=id+'_'+comp\n row_id.append(id_sample)\n target.append(preds_dict[comp][k])\n del df_test_img\n del X_test\n gc.collect()\n\ndf_sample = pd.DataFrame(\n {\n 'row_id': row_id,\n 'target':target\n },\n columns = ['row_id','target'] \n)\ndf_sample.to_csv('submission.csv',index=False)\n\nNotebook Exceeded Allowed Compute!!!", "votes": null }, { "id": "703372", "postDate": "12/26/2019 04:09:31", "content": "Well, the error says that the Notebook uses too much CPU\\RAM\\GPU. You might want to check the resource usage and make stuff more efficient. For example, use a smaller input size or use single channel models. Here's a good example of that. https://www.kaggle.com/c/bengaliai-cv19/discussion/123025
", "rawMarkdown": "Well, the error says that the Notebook uses too much CPU\\RAM\\GPU. You might want to check the resource usage and make stuff more efficient. For example, use a smaller input size or use single channel models. Here's a good example of that. https://www.kaggle.com/c/bengaliai-cv19/discussion/123025", "votes": null }, { "id": "703426", "postDate": "12/26/2019 06:27:44", "content": "The same thing happened with me recently while compiling a huge amount of data, I think it can be because of the outburst of the ram capacity.
", "rawMarkdown": "The same thing happened with me recently while compiling a huge amount of data, I think it can be because of the outburst of the ram capacity.", "votes": null }, { "id": "704061", "postDate": "12/27/2019 02:05:58", "content": "I saw the output file in my notebook, but when I scored, something went wrong！\nThe error is as follows ”Submission CSV Not Found”！\nIs this related to network connectivity?\n@drhabib @TahsinReasat \nThanks first!!!
", "rawMarkdown": "I saw the output file in my notebook, but when I scored, something went wrong！\nThe error is as follows ”Submission CSV Not Found”！\nIs this related to network connectivity?\n@drhabib @TahsinReasat \nThanks first!!!", "votes": null }, { "id": "704065", "postDate": "12/27/2019 02:21:19", "content": "It hard to say what can be error. One thing what you can do is share notebook... Pretty sure people will help you find what causing the error. Good Luck
", "rawMarkdown": "It hard to say what can be error. One thing what you can do is share notebook... Pretty sure people will help you find what causing the error. Good Luck", "votes": null }, { "id": "704075", "postDate": "12/27/2019 02:31:15", "content": "@drhabib @TahsinReasat Shared notebook https://www.kaggle.com/zhangeng/kernel716c21e272
", "rawMarkdown": "drhabib @TahsinReasat Shared notebook https://www.kaggle.com/zhangeng/kernel716c21e272", "votes": null }, { "id": "704272", "postDate": "12/27/2019 09:01:48", "content": "Hi Zhengeng, I see that you are training three separate models for the three components. Would it be possible for you to make an estimate on the inference time? \nYou can do that easily by changing the 'test' string to 'train' in the loaders. And use something like tqdm to give you an eta. Wrap it around range and change for i in range(4): to 'for i in tqdm(range(4)):'\nSince the number of train images and test images ar similar this would give a good enough estimate for the inference time.
I am getting this error while Submit to Competition\nNotebook Exceeded Allowed Compute
", "rawMarkdown": "I am getting this error while Submit to Competition\nNotebook Exceeded Allowed Compute", "votes": null }, { "id": "705148", "postDate": "12/28/2019 14:22:16", "content": "I am having the timeout problem too.\nHow can I change kaggle configuration to 1gpu and 2 cpu? To inference testing speed?
", "rawMarkdown": "I am having the timeout problem too.\nHow can I change kaggle configuration to 1gpu and 2 cpu? To inference testing speed?", "votes": null }, { "id": "706175", "postDate": "12/30/2019 02:59:53", "content": "Thanks. Finally, there are scores @drhabib @TahsinReasat
", "rawMarkdown": "Thanks. Finally, there are scores @drhabib @TahsinReasat", "votes": null }, { "id": "706178", "postDate": "12/30/2019 03:01:53", "content": "glad we could help! =) Good luck!
", "rawMarkdown": "glad we could help! =) Good luck!", "votes": null }, { "id": "733852", "postDate": "01/31/2020 15:30:48", "content": "I'm getting the same Notebook Exceeded Allowed Compute error message. I submitted a ticket to Kaggle.
", "rawMarkdown": "I'm getting the same Notebook Exceeded Allowed Compute error message. I submitted a ticket to Kaggle.", "votes": null }, { "id": "744287", "postDate": "02/12/2020 17:43:56", "content": "I've been having huge issues with this and I couldn't even submit 1's. So far I found that when I go to reshape the data, that seems to be the breaking point for me. I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help
", "rawMarkdown": "I've been having huge issues with this and I couldn't even submit 1's. So far I found that when I go to reshape the data, that seems to be the breaking point for me. I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help", "votes": null }, { "id": "744288", "postDate": "02/12/2020 17:44:42", "content": "UPDATE: I've been having huge issues with this and I couldn't even submit 1's. So far I found that when I go to reshape the data, that seems to be the breaking point for me. I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help
", "rawMarkdown": "**UPDATE:** I've been having huge issues with this and I couldn't even submit 1's. So far I found that when I go to reshape the data, that seems to be the breaking point for me. I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help", "votes": null } ], "comments": [ { "id": 702918, "author_name": "bitthal", "author_url": "", "post_date": "12/25/2019 10:59:33", "content": "facing same error
", "votes": null, "replies": [] }, { "id": 703089, "author_name": "drhabib", "author_url": "", "post_date": "12/25/2019 15:26:15", "content": "Ok here are few things that can go wrong:
\n\n1) Make sure you notebook is not connected to internet. You can disable this on the right side of the notebook \n
2) If you create local folders or other files make sure to remove them at the end of the notebook (except submission.csv): Below is the example how it should look like:
3) Make sure you have GPU hours left. Everyone gets 30h per week. You can check it by going to your profile and clicking \"edit profile\" and look for Quotes. Below is the example how it looks:\n
4) When you submit the test files get replaced by bigger dataset (e.g 200k images). You should make sure that your inference run in less then 9h. You can try to run prediction for 200000 images by replicating kaggle configuration 1 gpu and 2 cpu with 15gb ram.
\n\nGood luck =)
", "votes": null, "replies": [ { "id": 703308, "author_name": "zhangeng", "author_url": "", "post_date": "12/26/2019 01:38:14", "content": "\n", "votes": null, "replies": [] }, { "id": 703317, "author_name": "reasat", "author_url": "", "post_date": "12/26/2019 01:58:29", "content": "DrHB wrote:
\n \nOk here are few things that can go wrong:
\n \n1) Make sure you notebook is not connected to internet. You can disable this on the right side of the notebook \n
\n \n2) If you create local folders or other files make sure to remove them at the end of the notebook (except submission.csv): Below is the example how it should look like:
\n \n3) Make sure you have GPU hours left. Everyone gets 30h per week. You can check it by going to your profile and clicking \"edit profile\" and look for Quotes. Below is the example how it looks:\n
\n \n4) When you submit the test files get replaced by bigger dataset (e.g 200k images). You should make sure that your inference run in less then 9h. You can try to run prediction for 200000 images by replicating kaggle configuration 1 gpu and 2 cpu with 15gb ram.
\n \nGood luck =)
\n \n1.Thanks.\n 2.There are no problems in the first three points, I have done it.\n 3.Fourth, is there 200k test images?
\n
3) Yes, as stated in the data description page, the number of test images are roughly the same as the number of train images.
", "votes": null, "replies": [] }, { "id": 703337, "author_name": "zhangeng", "author_url": "", "post_date": "12/26/2019 02:51:08", "content": "@drhabib DrHB\nBecause I use the pre training model, which requires three channels, np..repeat (data, 3), the code is as follows:\n df_test_img = pd.read_parquet('/kaggle/input/bengaliai-cv19/test_image_data_{}.parquet'.format(i),dtype=np.uint) \n df_test_img.set_index('image_id', inplace=True)
\n\nX_test = resize(df_test_img, need_progress_bar=False)/255.0\nX_test = X_test.values.reshape(-1, reHEIGHT,reWIDTH)\nX_test = np.repeat(X_test,3)\nX_test = np.array(X_test).reshape(-1, reHEIGHT,reWIDTH, N_CHANNELS)\nHowever, the following error occurred: 《Notebook Exceeded Allowed Compute》
", "votes": null, "replies": [] }, { "id": 703363, "author_name": "zhangeng", "author_url": "", "post_date": "12/26/2019 03:37:55", "content": "@TahsinReasat\ncomponents = ['consonant_diacritic', 'grapheme_root', 'vowel_diacritic']\ntarget=[] # model predictions placeholder\nrow_id=[] # row_id place holder\nfor i in range(4):\n df_test_img = pd.read_parquet('/kaggle/input/bengaliai-cv19/test_image_data_{}.parquet'.format(i)) \n df_test_img.set_index('image_id', inplace=True)
\n\nX_test = resize(df_test_img, need_progress_bar=False)/255\n#X_test = X_test.values.reshape(-1, reHEIGHT,reWIDTH)\n#X_test = np.repeat(X_test,3)\nX_test = np.repeat(X_test.values,3).reshape(-1, reHEIGHT,reWIDTH, N_CHANNELS)\n\nfor pred in preds_dict:\n preds_dict[pred]=np.argmax(model_dict[pred].predict(X_test), axis=1)\n\nfor k,id in enumerate(df_test_img.index.values): \n for i,comp in enumerate(components):\n id_sample=id+'_'+comp\n row_id.append(id_sample)\n target.append(preds_dict[comp][k])\ndel df_test_img\ndel X_test\ngc.collect()\ndf_sample = pd.DataFrame(\n {\n 'row_id': row_id,\n 'target':target\n },\n columns = ['row_id','target'] \n)\ndf_sample.to_csv('submission.csv',index=False)
\n\nNotebook Exceeded Allowed Compute!!!
", "votes": null, "replies": [] }, { "id": 703372, "author_name": "reasat", "author_url": "", "post_date": "12/26/2019 04:09:31", "content": "Well, the error says that the Notebook uses too much CPU\\RAM\\GPU. You might want to check the resource usage and make stuff more efficient. For example, use a smaller input size or use single channel models. Here's a good example of that. https://www.kaggle.com/c/bengaliai-cv19/discussion/123025
", "votes": null, "replies": [] }, { "id": 705148, "author_name": "macarrony00", "author_url": "", "post_date": "12/28/2019 14:22:16", "content": "I am having the timeout problem too.\nHow can I change kaggle configuration to 1gpu and 2 cpu? To inference testing speed?
", "votes": null, "replies": [] } ] }, { "id": 703426, "author_name": "sabinhashmi", "author_url": "", "post_date": "12/26/2019 06:27:44", "content": "The same thing happened with me recently while compiling a huge amount of data, I think it can be because of the outburst of the ram capacity.
", "votes": null, "replies": [] }, { "id": 704061, "author_name": "zhangeng", "author_url": "", "post_date": "12/27/2019 02:05:58", "content": "I saw the output file in my notebook, but when I scored, something went wrong！\nThe error is as follows ”Submission CSV Not Found”！\nIs this related to network connectivity?\n@drhabib @TahsinReasat \nThanks first!!!
", "votes": null, "replies": [ { "id": 704065, "author_name": "drhabib", "author_url": "", "post_date": "12/27/2019 02:21:19", "content": "It hard to say what can be error. One thing what you can do is share notebook... Pretty sure people will help you find what causing the error. Good Luck
", "votes": null, "replies": [] }, { "id": 704075, "author_name": "zhangeng", "author_url": "", "post_date": "12/27/2019 02:31:15", "content": "@drhabib @TahsinReasat Shared notebook https://www.kaggle.com/zhangeng/kernel716c21e272
", "votes": null, "replies": [] }, { "id": 704272, "author_name": "reasat", "author_url": "", "post_date": "12/27/2019 09:01:48", "content": "Hi Zhengeng, I see that you are training three separate models for the three components. Would it be possible for you to make an estimate on the inference time? \nYou can do that easily by changing the 'test' string to 'train' in the loaders. And use something like tqdm to give you an eta. Wrap it around range and change for i in range(4): to 'for i in tqdm(range(4)):'\nSince the number of train images and test images ar similar this would give a good enough estimate for the inference time.
I am getting this error while Submit to Competition\nNotebook Exceeded Allowed Compute
", "votes": null, "replies": [ { "id": 744287, "author_name": "yeayates21", "author_url": "", "post_date": "02/12/2020 17:43:56", "content": "I've been having huge issues with this and I couldn't even submit 1's. So far I found that when I go to reshape the data, that seems to be the breaking point for me. I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help
", "votes": null, "replies": [] } ] }, { "id": 706175, "author_name": "zhangeng", "author_url": "", "post_date": "12/30/2019 02:59:53", "content": "Thanks. Finally, there are scores @drhabib @TahsinReasat
", "votes": null, "replies": [ { "id": 706178, "author_name": "drhabib", "author_url": "", "post_date": "12/30/2019 03:01:53", "content": "glad we could help! =) Good luck!
", "votes": null, "replies": [] } ] }, { "id": 733852, "author_name": "yeayates21", "author_url": "", "post_date": "01/31/2020 15:30:48", "content": "I'm getting the same Notebook Exceeded Allowed Compute error message. I submitted a ticket to Kaggle.
", "votes": null, "replies": [ { "id": 744288, "author_name": "yeayates21", "author_url": "", "post_date": "02/12/2020 17:44:42", "content": "UPDATE: I've been having huge issues with this and I couldn't even submit 1's. So far I found that when I go to reshape the data, that seems to be the breaking point for me. I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "702698": "Notebook is running normally\nbut ”Submit to Competition“ Error occurred\nNotebook Threw Exception or Notebook Timeout \nwhat's wrong???", "702918": "facing same error", "703089": "Ok here are few things that can go wrong:\n\n\n1) Make sure you notebook is not connected to internet. You can disable this on the right side of the notebook \n\n\n2) If you create local folders or other files make sure to remove them at the end of the notebook (except submission.csv): Below is the example how it should look like: \n\n3) Make sure you have GPU hours left. Everyone gets 30h per week. You can check it by going to your profile and clicking \"edit profile\" and look for Quotes. Below is the example how it looks:\n\n\n4) When you submit the test files get replaced by bigger dataset (e.g 200k images). You should make sure that your inference run in less then 9h. You can try to run prediction for 200000 images by replicating kaggle configuration 1 gpu and 2 cpu with 15gb ram.\n\nGood luck =)", "703308": "> **DrHB wrote:**\n> \n> Ok here are few things that can go wrong:\n> \n> \n> 1) Make sure you notebook is not connected to internet. You can disable this on the right side of the notebook \n> \n> \n> 2) If you create local folders or other files make sure to remove them at the end of the notebook (except submission.csv): Below is the example how it should look like: \n> \n> 3) Make sure you have GPU hours left. Everyone gets 30h per week. You can check it by going to your profile and clicking \"edit profile\" and look for Quotes. Below is the example how it looks:\n> \n> \n> 4) When you submit the test files get replaced by bigger dataset (e.g 200k images). You should make sure that your inference run in less then 9h. You can try to run prediction for 200000 images by replicating kaggle configuration 1 gpu and 2 cpu with 15gb ram.\n> \n> Good luck =) \n> \n1.Thanks.\n2.There are no problems in the first three points, I have done it.\n3.Fourth, is there 200k test images?", "703317": "3) Yes, as stated in the data description page, the number of test images are roughly the same as the number of train images.", "703337": "drhabib DrHB\nBecause I use the pre training model, which requires three channels, np..repeat (data, 3), the code is as follows:\n df_test_img = pd.read_parquet('/kaggle/input/bengaliai-cv19/test_image_data_{}.parquet'.format(i),dtype=np.uint) \n df_test_img.set_index('image_id', inplace=True)\n\n X_test = resize(df_test_img, need_progress_bar=False)/255.0\n X_test = X_test.values.reshape(-1, reHEIGHT,reWIDTH)\n X_test = np.repeat(X_test,3)\n X_test = np.array(X_test).reshape(-1, reHEIGHT,reWIDTH, N_CHANNELS)\nHowever, the following error occurred: 《Notebook Exceeded Allowed Compute》", "703363": "TahsinReasat\ncomponents = ['consonant_diacritic', 'grapheme_root', 'vowel_diacritic']\ntarget=[] # model predictions placeholder\nrow_id=[] # row_id place holder\nfor i in range(4):\n df_test_img = pd.read_parquet('/kaggle/input/bengaliai-cv19/test_image_data_{}.parquet'.format(i)) \n df_test_img.set_index('image_id', inplace=True)\n\n X_test = resize(df_test_img, need_progress_bar=False)/255\n #X_test = X_test.values.reshape(-1, reHEIGHT,reWIDTH)\n #X_test = np.repeat(X_test,3)\n X_test = np.repeat(X_test.values,3).reshape(-1, reHEIGHT,reWIDTH, N_CHANNELS)\n\n for pred in preds_dict:\n preds_dict[pred]=np.argmax(model_dict[pred].predict(X_test), axis=1)\n\n for k,id in enumerate(df_test_img.index.values): \n for i,comp in enumerate(components):\n id_sample=id+'_'+comp\n row_id.append(id_sample)\n target.append(preds_dict[comp][k])\n del df_test_img\n del X_test\n gc.collect()\n\ndf_sample = pd.DataFrame(\n {\n 'row_id': row_id,\n 'target':target\n },\n columns = ['row_id','target'] \n)\ndf_sample.to_csv('submission.csv',index=False)\n\nNotebook Exceeded Allowed Compute!!!", "703372": "Well, the error says that the Notebook uses too much CPU\\RAM\\GPU. You might want to check the resource usage and make stuff more efficient. For example, use a smaller input size or use single channel models. Here's a good example of that. https://www.kaggle.com/c/bengaliai-cv19/discussion/123025", "703426": "The same thing happened with me recently while compiling a huge amount of data, I think it can be because of the outburst of the ram capacity.", "704061": "I saw the output file in my notebook, but when I scored, something went wrong！\nThe error is as follows ”Submission CSV Not Found”！\nIs this related to network connectivity?\n@drhabib @TahsinReasat \nThanks first!!!", "704065": "It hard to say what can be error. One thing what you can do is share notebook... Pretty sure people will help you find what causing the error. Good Luck", "704075": "drhabib @TahsinReasat Shared notebook https://www.kaggle.com/zhangeng/kernel716c21e272", "704272": "Hi Zhengeng, I see that you are training three separate models for the three components. Would it be possible for you to make an estimate on the inference time? \nYou can do that easily by changing the 'test' string to 'train' in the loaders. And use something like tqdm to give you an eta. Wrap it around range and change `for i in range(4):` to 'for i in tqdm(range(4)):'\nSince the number of train images and test images ar similar this would give a good enough estimate for the inference time.", "704383": "I am getting this error while Submit to Competition\nNotebook Exceeded Allowed Compute", "705148": "I am having the timeout problem too.\nHow can I change kaggle configuration to 1gpu and 2 cpu? To inference testing speed?", "706175": "Thanks. Finally, there are scores @drhabib @TahsinReasat", "706178": "glad we could help! =) Good luck!", "733852": "I'm getting the same Notebook Exceeded Allowed Compute error message. I submitted a ticket to Kaggle.", "744287": "I've been having huge issues with this and I couldn't even submit 1's. So far I found that when I go to reshape the data, that seems to be the breaking point for me. I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help", "744288": "**UPDATE:** I've been having huge issues with this and I couldn't even submit 1's. So far I found that when I go to reshape the data, that seems to be the breaking point for me. I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help" }, "source": "meta" }