{
  "id": 145021,
  "title": "New to Machine Learning or to Kaggle? ",
  "url": "/competitions/prostate-cancer-grade-assessment/discussion/145021",
  "author_name": "Maggie",
  "post_date": "2020-04-21T16:43:52.124000",
  "votes": 9,
  "comment_count": 37,
  "views": 0,
  "content": "<p>New to machine learning and data science? No question is too basic or too simple. Feel free to start your own thread, or use this thread as a place to post any first-timer clarifying questions for the Kaggle community to help you with! </p>\n\n<p>If you would consider yourself a beginner but don't know where to get started, let other Kagglers help you take your first steps here!</p>\n\n<p>New to Kaggle? Take a look at a few videos our very own Dr. Rachael Tatman has put together to learn a bit more about <a href=\"https://www.youtube.com/watch?v=aIus8si_Et0\">site etiquette</a>, <a href=\"https://www.youtube.com/watch?v=sEJHyuWKd-s\">Kaggle lingo</a>, and <a href=\"https://www.youtube.com/watch?&amp;v=GJBOMWpLpTQ\">how to enter a competition using Kaggle Notebooks</a>.</p>",
  "messages": [
    {
      "id": 815572,
      "postDate": "2020-04-21T16:43:52.123Z",
      "content": "<p>New to machine learning and data science? No question is too basic or too simple. Feel free to start your own thread, or use this thread as a place to post any first-timer clarifying questions for the Kaggle community to help you with! </p>\n\n<p>If you would consider yourself a beginner but don't know where to get started, let other Kagglers help you take your first steps here!</p>\n\n<p>New to Kaggle? Take a look at a few videos our very own Dr. Rachael Tatman has put together to learn a bit more about <a href=\"https://www.youtube.com/watch?v=aIus8si_Et0\">site etiquette</a>, <a href=\"https://www.youtube.com/watch?v=sEJHyuWKd-s\">Kaggle lingo</a>, and <a href=\"https://www.youtube.com/watch?&amp;v=GJBOMWpLpTQ\">how to enter a competition using Kaggle Notebooks</a>.</p>",
      "rawMarkdown": "New to machine learning and data science? No question is too basic or too simple. Feel free to start your own thread, or use this thread as a place to post any first-timer clarifying questions for the Kaggle community to help you with! \n\nIf you would consider yourself a beginner but don't know where to get started, let other Kagglers help you take your first steps here!\n\nNew to Kaggle? Take a look at a few videos our very own Dr. Rachael Tatman has put together to learn a bit more about [site etiquette](https://www.youtube.com/watch?v=aIus8si_Et0), [Kaggle lingo](https://www.youtube.com/watch?v=sEJHyuWKd-s), and [how to enter a competition using Kaggle Notebooks](https://www.youtube.com/watch?&amp;v=GJBOMWpLpTQ).",
      "votes": 9
    },
    {
      "id": 819329,
      "postDate": "2020-04-24T14:13:22.223Z",
      "content": "<p>Hello, new to Kaggle. Is training of 193GB of data possible in Kaggle Notebooks?</p>",
      "rawMarkdown": "Hello, new to Kaggle. Is training of 193GB of data possible in Kaggle Notebooks?",
      "votes": 8
    },
    {
      "id": 921237,
      "postDate": "2020-07-09T06:34:45.020Z",
      "content": "<p>I think most use local resources, but I make do with just the kaggle. It's quite a challenge to make progress, but it is possible. </p>",
      "rawMarkdown": "I think most use local resources, but I make do with just the kaggle. It's quite a challenge to make progress, but it is possible. ",
      "votes": 1
    },
    {
      "id": 905541,
      "postDate": "2020-06-28T16:17:57.913Z",
      "content": "<p>Does anyone know how to handle \"Submission csv not found\"???????? The csv was named 'submission.csv' and the internet was turned off as mentioned in the description of the competition. Is there something that I am missing? Thanks in advance.</p>",
      "rawMarkdown": "Does anyone know how to handle \"Submission csv not found\"???????? The csv was named 'submission.csv' and the internet was turned off as mentioned in the description of the competition. Is there something that I am missing? Thanks in advance.\n",
      "votes": 1
    },
    {
      "id": 853024,
      "postDate": "2020-05-18T22:35:56.277Z",
      "content": "<p>Hi, this is my first Kaggle ever! I'm following this guide on downloading from kaggle into colab: <a href=\"https://www.kaggle.com/general/6604\">https://www.kaggle.com/general/6604</a></p>\n\n<p>However, I'm noticing that the ~30 zipfiles of .tiff.zip in my folder all hold only one image inside, while there should be a total of ~10,000. Could anyone please advise as to what I might be doing wrong? Thank you!</p>",
      "rawMarkdown": "Hi, this is my first Kaggle ever! I'm following this guide on downloading from kaggle into colab: https://www.kaggle.com/general/6604\n\nHowever, I'm noticing that the ~30 zipfiles of .tiff.zip in my folder all hold only one image inside, while there should be a total of ~10,000. Could anyone please advise as to what I might be doing wrong? Thank you!",
      "votes": 1
    },
    {
      "id": 842122,
      "postDate": "2020-05-11T07:59:49.573Z",
      "content": "<p>I have a very basic question. The kernels all load test images from the directory 'kaggle/input/prostate-cancer-grade-assessment/test_images'.</p>\n\n<p>But when I list the directories I get: \n/kaggle/input\n/kaggle/input/densenet-keras\n/kaggle/input/prostate-cancer-grade-assessment\n/kaggle/input/prostate-cancer-grade-assessment/train_images\n/kaggle/input/prostate-cancer-grade-assessment/train_label_masks</p>\n\n<p>No test_images directory. When I try to load them, of course this fails. </p>",
      "rawMarkdown": "I have a very basic question. The kernels all load test images from the directory 'kaggle/input/prostate-cancer-grade-assessment/test_images'.\n\nBut when I list the directories I get: \n/kaggle/input\n/kaggle/input/densenet-keras\n/kaggle/input/prostate-cancer-grade-assessment\n/kaggle/input/prostate-cancer-grade-assessment/train_images\n/kaggle/input/prostate-cancer-grade-assessment/train_label_masks\n\nNo test_images directory. When I try to load them, of course this fails. ",
      "votes": 1,
      "replies": [
        {
          "id": 842296,
          "postDate": "2020-05-11T10:08:36.840Z",
          "content": "<p>Ah, I see. Have to commit the kernel to get access to the test directory. </p>",
          "rawMarkdown": "Ah, I see. Have to commit the kernel to get access to the test directory. ",
          "votes": 1
        }
      ]
    },
    {
      "id": 945204,
      "postDate": "2020-07-25T16:40:51.647Z",
      "content": "<p>Thanks for sharing this. I am starting today and hoping to able to join good team!</p>",
      "rawMarkdown": "Thanks for sharing this. I am starting today and hoping to able to join good team!\n"
    },
    {
      "id": 921218,
      "postDate": "2020-07-09T06:14:57.920Z",
      "content": "<p>I'm new to kaggle. Is one GPU with 9 hours per session and 16GB memory is really enough for competitors?\nOr does most use other local/cloud resources?</p>",
      "rawMarkdown": "I'm new to kaggle. Is one GPU with 9 hours per session and 16GB memory is really enough for competitors?\nOr does most use other local/cloud resources?"
    },
    {
      "id": 892805,
      "postDate": "2020-06-19T07:11:06.223Z",
      "content": "<p>Hi I,m new to Kaggle...struggling to grasp the concept of AI </p>",
      "rawMarkdown": "Hi I,m new to Kaggle...struggling to grasp the concept of AI ",
      "replies": [
        {
          "id": 900272,
          "postDate": "2020-06-24T18:00:28.880Z",
          "content": "<p>Check out our learn courses that can walk you through the basics. </p>",
          "rawMarkdown": "Check out our learn courses that can walk you through the basics. ",
          "votes": 1
        }
      ]
    },
    {
      "id": 892520,
      "postDate": "2020-06-19T00:49:20.533Z",
      "content": "<p>Hi I'm new to python and deep learning.\nWhen we resize the image, did we not lose a bunch of efficient data?\nOr it's necessary due to computation cost?</p>",
      "rawMarkdown": "Hi I'm new to python and deep learning.\nWhen we resize the image, did we not lose a bunch of efficient data?\nOr it's necessary due to computation cost?",
      "replies": [
        {
          "id": 892619,
          "postDate": "2020-06-19T03:37:31.263Z",
          "content": "<p>We would like to use the largest image possible. Only downsize to make things work with the computation available. In this competition, has to work within what is available to the kernel with gpu. </p>",
          "rawMarkdown": "We would like to use the largest image possible. Only downsize to make things work with the computation available. In this competition, has to work within what is available to the kernel with gpu. "
        },
        {
          "id": 892720,
          "postDate": "2020-06-19T05:32:30.687Z",
          "content": "<p>Thanks Andy! That's what I was wondering, just to making sure. </p>",
          "rawMarkdown": "Thanks Andy! That's what I was wondering, just to making sure. "
        }
      ]
    },
    {
      "id": 843351,
      "postDate": "2020-05-12T01:59:24.850Z",
      "content": "<p>In the datasets part on what topics do i collect data? and how to apprach any data collection ask</p>",
      "rawMarkdown": "In the datasets part on what topics do i collect data? and how to apprach any data collection ask"
    },
    {
      "id": 842295,
      "postDate": "2020-05-11T10:07:43.277Z",
      "content": "<p>Hello, I like ML algorithms and mathematics behind them. But when I started working on my first data science hackathon, I experienced that there is more of feature engineering, converting categorical feature into numerical data and thinking about which method(one hot encoding/mean encoding etc) to use for it. And for training part you use sklearn, that's it. I don't like so called data pre-processing part at all. So is this field not for me? And Is the work demand in this role in firms similar to what people do in hackathons? </p>",
      "rawMarkdown": "Hello, I like ML algorithms and mathematics behind them. But when I started working on my first data science hackathon, I experienced that there is more of feature engineering, converting categorical feature into numerical data and thinking about which method(one hot encoding/mean encoding etc) to use for it. And for training part you use sklearn, that's it. I don't like so called data pre-processing part at all. So is this field not for me? And Is the work demand in this role in firms similar to what people do in hackathons? ",
      "replies": [
        {
          "id": 842688,
          "postDate": "2020-05-11T15:04:32.780Z",
          "content": "<p>If you work in deep learning there is usually more applied research work to do. Even if you don't design the algorithm making the application of a new paper is pretty common so it's more work than just .fit .predict with sklearn. And obviously researchers spend more time on the fundamentals.</p>\n\n<p>In industry you also spend a lot of time doing cleaning, designing datasets etc. Starting with a project with an already well defined supervised learning setup never happens.</p>",
          "rawMarkdown": "If you work in deep learning there is usually more applied research work to do. Even if you don't design the algorithm making the application of a new paper is pretty common so it's more work than just .fit .predict with sklearn. And obviously researchers spend more time on the fundamentals.\n\nIn industry you also spend a lot of time doing cleaning, designing datasets etc. Starting with a project with an already well defined supervised learning setup never happens.",
          "votes": 2
        }
      ]
    },
    {
      "id": 839515,
      "postDate": "2020-05-09T12:41:00.237Z",
      "content": "<p>is it possible to get more than 16GB of RAM on kaggle? </p>",
      "rawMarkdown": "is it possible to get more than 16GB of RAM on kaggle? ",
      "replies": [
        {
          "id": 874279,
          "postDate": "2020-06-04T19:30:04.633Z",
          "content": "<p>Apparently, the answer is no.\nYou will find the answer here: <a href=\"https://www.kaggle.com/docs/notebooks#technical-specifications\">Technical Specifications</a></p>",
          "rawMarkdown": "Apparently, the answer is no.\nYou will find the answer here: [Technical Specifications](https://www.kaggle.com/docs/notebooks#technical-specifications)"
        }
      ]
    },
    {
      "id": 818411,
      "postDate": "2020-04-23T21:18:58.177Z",
      "content": "<p>Hello!</p>\n\n<p>I don't get how your solution is evaluated on a private dataset when you submit it. \nDo you even submit a file or a notebook?</p>",
      "rawMarkdown": "Hello!\n\nI don't get how your solution is evaluated on a private dataset when you submit it. \nDo you even submit a file or a notebook?"
    },
    {
      "id": 816545,
      "postDate": "2020-04-22T12:12:53.323Z",
      "content": "<p>Hello i have one doubt!!In competitions i saw many people share their notebooks ?Can we take help from these notebooks is that allowed in the competition</p>",
      "rawMarkdown": "Hello i have one doubt!!In competitions i saw many people share their notebooks ?Can we take help from these notebooks is that allowed in the competition",
      "replies": [
        {
          "id": 818221,
          "postDate": "2020-04-23T17:51:07.663Z",
          "content": "<p>Yeah. Many people will share their work in the form of a public notebook. You can fork from these ideas and try to improve them yourself.</p>",
          "rawMarkdown": "Yeah. Many people will share their work in the form of a public notebook. You can fork from these ideas and try to improve them yourself.",
          "votes": 2
        },
        {
          "id": 818769,
          "postDate": "2020-04-24T05:40:03.647Z",
          "content": "<p><a href=\"/matthewmasters\">@matthewmasters</a> \nThanks a lot i am a begginer will try to do that!!!\nthanks a lot again :) </p>",
          "rawMarkdown": "@matthewmasters \nThanks a lot i am a begginer will try to do that!!!\nthanks a lot again :) \n"
        },
        {
          "id": 818832,
          "postDate": "2020-04-24T06:28:23.750Z",
          "content": "<p><a href=\"/matthewmasters\">@matthewmasters</a> actually i was thinking if people share their work wont the score be same?</p>",
          "rawMarkdown": "@matthewmasters actually i was thinking if people share their work wont the score be same?"
        },
        {
          "id": 838866,
          "postDate": "2020-05-08T21:05:00.080Z",
          "content": "<p>Not all work is shared. Leaders at best will give you a base without the many tricks they use to win (until the competition is over at least). They may share details on the discussions though. A good practice for beginners is to build (or copy) a base and then look at additional notebooks and discussions to implement the ideas proposed and keep the one that give some improvement.</p>",
          "rawMarkdown": "Not all work is shared. Leaders at best will give you a base without the many tricks they use to win (until the competition is over at least). They may share details on the discussions though. A good practice for beginners is to build (or copy) a base and then look at additional notebooks and discussions to implement the ideas proposed and keep the one that give some improvement.",
          "votes": 2
        }
      ]
    },
    {
      "id": 815842,
      "postDate": "2020-04-21T21:58:30.187Z",
      "content": "<p>Hello! I have some trouble in extracting data, specifically all images from the .tiff multilevel files in Mac OS, I am new to this approach, any recommendations or good practices in order to extract all of the data needed for the model? Thanks in advance!</p>",
      "rawMarkdown": "Hello! I have some trouble in extracting data, specifically all images from the .tiff multilevel files in Mac OS, I am new to this approach, any recommendations or good practices in order to extract all of the data needed for the model? Thanks in advance!",
      "replies": [
        {
          "id": 815851,
          "postDate": "2020-04-21T22:08:46.880Z",
          "content": "<p>Wouter covers this in <a href=\"https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset\">the starter notebook</a>.</p>",
          "rawMarkdown": "Wouter covers this in [the starter notebook](https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset).",
          "votes": 1
        },
        {
          "id": 815871,
          "postDate": "2020-04-21T22:27:21.950Z",
          "content": "<p>Worked for me, thank you!</p>",
          "rawMarkdown": "Worked for me, thank you!"
        },
        {
          "id": 816640,
          "postDate": "2020-04-22T13:37:54.033Z",
          "content": "<p>I got some problems when trying to access \nTroubleshooting Info\nUser:amenyhmidi@gmail.com\nURL:<a href=\"https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset\">https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset</a></p>",
          "rawMarkdown": "I got some problems when trying to access \nTroubleshooting Info\nUser:amenyhmidi@gmail.com\nURL:https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset",
          "votes": 1
        },
        {
          "id": 816760,
          "postDate": "2020-04-22T15:06:20.137Z",
          "rawMarkdown": ""
        },
        {
          "id": 817197,
          "postDate": "2020-04-22T23:50:06.220Z",
          "content": "<p>I had this problem as well, but if you go to thread titled \"A warm welcome from the organizer team,\" there is a working link to the notebook.</p>",
          "rawMarkdown": "I had this problem as well, but if you go to thread titled \"A warm welcome from the organizer team,\" there is a working link to the notebook."
        },
        {
          "id": 818058,
          "postDate": "2020-04-23T15:54:17.330Z",
          "content": "<p>Sorry about that, I've fixed my link. This is the valid one: <a href=\"https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset\">https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset</a></p>",
          "rawMarkdown": "Sorry about that, I've fixed my link. This is the valid one: https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset"
        }
      ]
    },
    {
      "id": 903683,
      "postDate": "2020-06-27T03:38:00.477Z",
      "content": "<p>Thanks a lot for this video links</p>",
      "rawMarkdown": "Thanks a lot for this video links"
    },
    {
      "id": 902153,
      "postDate": "2020-06-26T00:45:19.383Z",
      "content": "<p>Thanks!</p>",
      "rawMarkdown": "Thanks!"
    },
    {
      "id": 892849,
      "postDate": "2020-06-19T07:48:05.233Z",
      "content": "<p>Thanks it really helps </p>",
      "rawMarkdown": "Thanks it really helps "
    },
    {
      "id": 849786,
      "postDate": "2020-05-16T04:21:57.350Z",
      "content": "<p>Thank you!</p>",
      "rawMarkdown": "Thank you!"
    },
    {
      "id": 840745,
      "postDate": "2020-05-10T10:39:25.060Z",
      "content": "<p>Thank you.</p>",
      "rawMarkdown": "Thank you."
    },
    {
      "id": 839277,
      "postDate": "2020-05-09T08:31:09.627Z",
      "content": "<p>Thank you so much for your help.</p>",
      "rawMarkdown": "Thank you so much for your help."
    }
  ],
  "comments": [
    {
      "id": 819329,
      "author_name": "Himanshu Agarwal",
      "author_url": "",
      "post_date": "2020-04-24T14:13:22.223000",
      "content": "<p>Hello, new to Kaggle. Is training of 193GB of data possible in Kaggle Notebooks?</p>",
      "votes": 8,
      "replies": []
    },
    {
      "id": 921237,
      "author_name": "andy jennings",
      "author_url": "",
      "post_date": "2020-07-09T06:34:45.020000",
      "content": "<p>I think most use local resources, but I make do with just the kaggle. It's quite a challenge to make progress, but it is possible. </p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 905541,
      "author_name": "Vibeesh",
      "author_url": "",
      "post_date": "2020-06-28T16:17:57.913000",
      "content": "<p>Does anyone know how to handle \"Submission csv not found\"???????? The csv was named 'submission.csv' and the internet was turned off as mentioned in the description of the competition. Is there something that I am missing? Thanks in advance.</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 853024,
      "author_name": "Jonathan Ko",
      "author_url": "",
      "post_date": "2020-05-18T22:35:56.277000",
      "content": "<p>Hi, this is my first Kaggle ever! I'm following this guide on downloading from kaggle into colab: <a href=\"https://www.kaggle.com/general/6604\">https://www.kaggle.com/general/6604</a></p>\n\n<p>However, I'm noticing that the ~30 zipfiles of .tiff.zip in my folder all hold only one image inside, while there should be a total of ~10,000. Could anyone please advise as to what I might be doing wrong? Thank you!</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 842122,
      "author_name": "andy jennings",
      "author_url": "",
      "post_date": "2020-05-11T07:59:49.573000",
      "content": "<p>I have a very basic question. The kernels all load test images from the directory 'kaggle/input/prostate-cancer-grade-assessment/test_images'.</p>\n\n<p>But when I list the directories I get: \n/kaggle/input\n/kaggle/input/densenet-keras\n/kaggle/input/prostate-cancer-grade-assessment\n/kaggle/input/prostate-cancer-grade-assessment/train_images\n/kaggle/input/prostate-cancer-grade-assessment/train_label_masks</p>\n\n<p>No test_images directory. When I try to load them, of course this fails. </p>",
      "votes": 1,
      "replies": [
        {
          "id": 842296,
          "author_name": "andy jennings",
          "author_url": "",
          "post_date": "2020-05-11T10:08:36.840000",
          "content": "<p>Ah, I see. Have to commit the kernel to get access to the test directory. </p>",
          "votes": 1,
          "replies": []
        }
      ]
    },
    {
      "id": 945204,
      "author_name": "Saatvik Aggarwal",
      "author_url": "",
      "post_date": "2020-07-25T16:40:51.647000",
      "content": "<p>Thanks for sharing this. I am starting today and hoping to able to join good team!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 921218,
      "author_name": "Tomer Sidi",
      "author_url": "",
      "post_date": "2020-07-09T06:14:57.920000",
      "content": "<p>I'm new to kaggle. Is one GPU with 9 hours per session and 16GB memory is really enough for competitors?\nOr does most use other local/cloud resources?</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 892805,
      "author_name": "Rajesh Kr. Ranjan",
      "author_url": "",
      "post_date": "2020-06-19T07:11:06.223000",
      "content": "<p>Hi I,m new to Kaggle...struggling to grasp the concept of AI </p>",
      "votes": 0,
      "replies": [
        {
          "id": 900272,
          "author_name": "Maggie",
          "author_url": "",
          "post_date": "2020-06-24T18:00:28.880000",
          "content": "<p>Check out our learn courses that can walk you through the basics. </p>",
          "votes": 1,
          "replies": []
        }
      ]
    },
    {
      "id": 892520,
      "author_name": "Goob Doy",
      "author_url": "",
      "post_date": "2020-06-19T00:49:20.533000",
      "content": "<p>Hi I'm new to python and deep learning.\nWhen we resize the image, did we not lose a bunch of efficient data?\nOr it's necessary due to computation cost?</p>",
      "votes": 0,
      "replies": [
        {
          "id": 892619,
          "author_name": "andy jennings",
          "author_url": "",
          "post_date": "2020-06-19T03:37:31.263000",
          "content": "<p>We would like to use the largest image possible. Only downsize to make things work with the computation available. In this competition, has to work within what is available to the kernel with gpu. </p>",
          "votes": 0,
          "replies": []
        },
        {
          "id": 892720,
          "author_name": "Goob Doy",
          "author_url": "",
          "post_date": "2020-06-19T05:32:30.687000",
          "content": "<p>Thanks Andy! That's what I was wondering, just to making sure. </p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 843351,
      "author_name": "",
      "author_url": "",
      "post_date": "2020-05-12T01:59:24.850000",
      "content": "<p>In the datasets part on what topics do i collect data? and how to apprach any data collection ask</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 842295,
      "author_name": "iTrust",
      "author_url": "",
      "post_date": "2020-05-11T10:07:43.277000",
      "content": "<p>Hello, I like ML algorithms and mathematics behind them. But when I started working on my first data science hackathon, I experienced that there is more of feature engineering, converting categorical feature into numerical data and thinking about which method(one hot encoding/mean encoding etc) to use for it. And for training part you use sklearn, that's it. I don't like so called data pre-processing part at all. So is this field not for me? And Is the work demand in this role in firms similar to what people do in hackathons? </p>",
      "votes": 0,
      "replies": [
        {
          "id": 842688,
          "author_name": "Arnaud Roussel",
          "author_url": "",
          "post_date": "2020-05-11T15:04:32.780000",
          "content": "<p>If you work in deep learning there is usually more applied research work to do. Even if you don't design the algorithm making the application of a new paper is pretty common so it's more work than just .fit .predict with sklearn. And obviously researchers spend more time on the fundamentals.</p>\n\n<p>In industry you also spend a lot of time doing cleaning, designing datasets etc. Starting with a project with an already well defined supervised learning setup never happens.</p>",
          "votes": 2,
          "replies": []
        }
      ]
    },
    {
      "id": 839515,
      "author_name": "Gershinen Shanding",
      "author_url": "",
      "post_date": "2020-05-09T12:41:00.237000",
      "content": "<p>is it possible to get more than 16GB of RAM on kaggle? </p>",
      "votes": 0,
      "replies": [
        {
          "id": 874279,
          "author_name": "jabertuhin",
          "author_url": "",
          "post_date": "2020-06-04T19:30:04.633000",
          "content": "<p>Apparently, the answer is no.\nYou will find the answer here: <a href=\"https://www.kaggle.com/docs/notebooks#technical-specifications\">Technical Specifications</a></p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 818411,
      "author_name": "Georgy Ginzburg",
      "author_url": "",
      "post_date": "2020-04-23T21:18:58.177000",
      "content": "<p>Hello!</p>\n\n<p>I don't get how your solution is evaluated on a private dataset when you submit it. \nDo you even submit a file or a notebook?</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 816545,
      "author_name": "NaturalNeuralNetwork",
      "author_url": "",
      "post_date": "2020-04-22T12:12:53.323000",
      "content": "<p>Hello i have one doubt!!In competitions i saw many people share their notebooks ?Can we take help from these notebooks is that allowed in the competition</p>",
      "votes": 0,
      "replies": [
        {
          "id": 818221,
          "author_name": "Matt",
          "author_url": "",
          "post_date": "2020-04-23T17:51:07.663000",
          "content": "<p>Yeah. Many people will share their work in the form of a public notebook. You can fork from these ideas and try to improve them yourself.</p>",
          "votes": 2,
          "replies": []
        },
        {
          "id": 818769,
          "author_name": "NaturalNeuralNetwork",
          "author_url": "",
          "post_date": "2020-04-24T05:40:03.647000",
          "content": "<p><a href=\"/matthewmasters\">@matthewmasters</a> \nThanks a lot i am a begginer will try to do that!!!\nthanks a lot again :) </p>",
          "votes": 0,
          "replies": []
        },
        {
          "id": 818832,
          "author_name": "NaturalNeuralNetwork",
          "author_url": "",
          "post_date": "2020-04-24T06:28:23.750000",
          "content": "<p><a href=\"/matthewmasters\">@matthewmasters</a> actually i was thinking if people share their work wont the score be same?</p>",
          "votes": 0,
          "replies": []
        },
        {
          "id": 838866,
          "author_name": "Arnaud Roussel",
          "author_url": "",
          "post_date": "2020-05-08T21:05:00.080000",
          "content": "<p>Not all work is shared. Leaders at best will give you a base without the many tricks they use to win (until the competition is over at least). They may share details on the discussions though. A good practice for beginners is to build (or copy) a base and then look at additional notebooks and discussions to implement the ideas proposed and keep the one that give some improvement.</p>",
          "votes": 2,
          "replies": []
        }
      ]
    },
    {
      "id": 815842,
      "author_name": "Freddy Maldonado",
      "author_url": "",
      "post_date": "2020-04-21T21:58:30.187000",
      "content": "<p>Hello! I have some trouble in extracting data, specifically all images from the .tiff multilevel files in Mac OS, I am new to this approach, any recommendations or good practices in order to extract all of the data needed for the model? Thanks in advance!</p>",
      "votes": 0,
      "replies": [
        {
          "id": 815851,
          "author_name": "Sohier Dane",
          "author_url": "",
          "post_date": "2020-04-21T22:08:46.880000",
          "content": "<p>Wouter covers this in <a href=\"https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset\">the starter notebook</a>.</p>",
          "votes": 1,
          "replies": []
        },
        {
          "id": 815871,
          "author_name": "Freddy Maldonado",
          "author_url": "",
          "post_date": "2020-04-21T22:27:21.950000",
          "content": "<p>Worked for me, thank you!</p>",
          "votes": 0,
          "replies": []
        },
        {
          "id": 816640,
          "author_name": "AmeniHMD",
          "author_url": "",
          "post_date": "2020-04-22T13:37:54.033000",
          "content": "<p>I got some problems when trying to access \nTroubleshooting Info\nUser:amenyhmidi@gmail.com\nURL:<a href=\"https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset\">https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset</a></p>",
          "votes": 1,
          "replies": []
        },
        {
          "id": 816760,
          "author_name": "Al",
          "author_url": "",
          "post_date": "2020-04-22T15:06:20.137000",
          "content": "",
          "votes": 0,
          "replies": []
        },
        {
          "id": 817197,
          "author_name": "George Yacu",
          "author_url": "",
          "post_date": "2020-04-22T23:50:06.220000",
          "content": "<p>I had this problem as well, but if you go to thread titled \"A warm welcome from the organizer team,\" there is a working link to the notebook.</p>",
          "votes": 0,
          "replies": []
        },
        {
          "id": 818058,
          "author_name": "Sohier Dane",
          "author_url": "",
          "post_date": "2020-04-23T15:54:17.330000",
          "content": "<p>Sorry about that, I've fixed my link. This is the valid one: <a href=\"https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset\">https://www.kaggle.com/wouterbulten/getting-started-with-the-panda-dataset</a></p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 903683,
      "author_name": "Salman Ibne Eunus",
      "author_url": "",
      "post_date": "2020-06-27T03:38:00.477000",
      "content": "<p>Thanks a lot for this video links</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 902153,
      "author_name": "Rashidul Hasan Hridoy",
      "author_url": "",
      "post_date": "2020-06-26T00:45:19.383000",
      "content": "<p>Thanks!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 892849,
      "author_name": "Shrirang Mhalgi",
      "author_url": "",
      "post_date": "2020-06-19T07:48:05.233000",
      "content": "<p>Thanks it really helps </p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 849786,
      "author_name": "Aaditya Diwan",
      "author_url": "",
      "post_date": "2020-05-16T04:21:57.350000",
      "content": "<p>Thank you!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 840745,
      "author_name": "Akash Dey",
      "author_url": "",
      "post_date": "2020-05-10T10:39:25.060000",
      "content": "<p>Thank you.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 839277,
      "author_name": "Ritika017",
      "author_url": "",
      "post_date": "2020-05-09T08:31:09.627000",
      "content": "<p>Thank you so much for your help.</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "815572": "New to machine learning and data science? No question is too basic or too simple. Feel free to start your own thread, or use this thread as a place to post any first-timer clarifying questions for the Kaggle community to help you with! \n\nIf you would consider yourself a beginner but don't know where to get started, let other Kagglers help you take your first steps here!\n\nNew to Kaggle? Take a look at a few videos our very own Dr. Rachael Tatman has put together to learn a bit more about [site etiquette](https://www.youtube.com/watch?v=aIus8si_Et0), [Kaggle lingo](https://www.youtube.com/watch?v=sEJHyuWKd-s), and [how to enter a competition using Kaggle Notebooks](https://www.youtube.com/watch?&amp;v=GJBOMWpLpTQ).",
    "819329": "Hello, new to Kaggle. Is training of 193GB of data possible in Kaggle Notebooks?",
    "921237": "I think most use local resources, but I make do with just the kaggle. It's quite a challenge to make progress, but it is possible. ",
    "905541": "Does anyone know how to handle \"Submission csv not found\"???????? The csv was named 'submission.csv' and the internet was turned off as mentioned in the description of the competition. Is there something that I am missing? Thanks in advance.\n",
    "853024": "Hi, this is my first Kaggle ever! I'm following this guide on downloading from kaggle into colab: https://www.kaggle.com/general/6604\n\nHowever, I'm noticing that the ~30 zipfiles of .tiff.zip in my folder all hold only one image inside, while there should be a total of ~10,000. Could anyone please advise as to what I might be doing wrong? Thank you!",
    "842122": "I have a very basic question. The kernels all load test images from the directory 'kaggle/input/prostate-cancer-grade-assessment/test_images'.\n\nBut when I list the directories I get: \n/kaggle/input\n/kaggle/input/densenet-keras\n/kaggle/input/prostate-cancer-grade-assessment\n/kaggle/input/prostate-cancer-grade-assessment/train_images\n/kaggle/input/prostate-cancer-grade-assessment/train_label_masks\n\nNo test_images directory. When I try to load them, of course this fails. ",
    "945204": "Thanks for sharing this. I am starting today and hoping to able to join good team!\n",
    "921218": "I'm new to kaggle. Is one GPU with 9 hours per session and 16GB memory is really enough for competitors?\nOr does most use other local/cloud resources?",
    "892805": "Hi I,m new to Kaggle...struggling to grasp the concept of AI ",
    "892520": "Hi I'm new to python and deep learning.\nWhen we resize the image, did we not lose a bunch of efficient data?\nOr it's necessary due to computation cost?",
    "843351": "In the datasets part on what topics do i collect data? and how to apprach any data collection ask",
    "842295": "Hello, I like ML algorithms and mathematics behind them. But when I started working on my first data science hackathon, I experienced that there is more of feature engineering, converting categorical feature into numerical data and thinking about which method(one hot encoding/mean encoding etc) to use for it. And for training part you use sklearn, that's it. I don't like so called data pre-processing part at all. So is this field not for me? And Is the work demand in this role in firms similar to what people do in hackathons? ",
    "839515": "is it possible to get more than 16GB of RAM on kaggle? ",
    "818411": "Hello!\n\nI don't get how your solution is evaluated on a private dataset when you submit it. \nDo you even submit a file or a notebook?",
    "816545": "Hello i have one doubt!!In competitions i saw many people share their notebooks ?Can we take help from these notebooks is that allowed in the competition",
    "815842": "Hello! I have some trouble in extracting data, specifically all images from the .tiff multilevel files in Mac OS, I am new to this approach, any recommendations or good practices in order to extract all of the data needed for the model? Thanks in advance!",
    "903683": "Thanks a lot for this video links",
    "902153": "Thanks!",
    "892849": "Thanks it really helps ",
    "849786": "Thank you!",
    "840745": "Thank you.",
    "839277": "Thank you so much for your help."
  }
}