{ "id": 202467, "title": "Learning path starts with questions?", "url": "/competitions/riiid-test-answer-prediction/discussion/202467", "author_name": "", "post_date": "2020-12-10T06:36:57.040360100Z", "votes": null, "comment_count": 6, "views": 0, "content": "

Hello to everyone,
\nI've started analysing learning paths for randomly selected users and found that for all questions go before lectures? Why it is so? I'd expected lectures first then questions then next cycle.

", "messages": [ { "id": "1107985", "postDate": "12/10/2020 06:36:57", "content": "

Hello to everyone,
\nI've started analysing learning paths for randomly selected users and found that for all questions go before lectures? Why it is so? I'd expected lectures first then questions then next cycle.

", "rawMarkdown": "Hello to everyone,\nI've started analysing learning paths for randomly selected users and found that for all questions go before lectures? Why it is so? I'd expected lectures first then questions then next cycle.", "votes": null }, { "id": "1108238", "postDate": "12/10/2020 12:52:57", "content": "

That is not the case - I just checked a user, user_id ==705741139.
\nDid you concat your questions and lectures dataframes - in which case you will see lectures at the end.

", "rawMarkdown": "That is not the case - I just checked a user, user_id ==705741139. \nDid you concat your questions and lectures dataframes - in which case you will see lectures at the end.", "votes": null }, { "id": "1108243", "postDate": "12/10/2020 13:02:18", "content": "

Thank you for answering. No I did not concatenated. I've just explored train df. Let's choose user 115. Head of its df ordered by timestamp:
\n timestamp user_id content_id content_type_id answered_correctly
\n0 0 115 5692 0 1
\n1 56943 115 5716 0 1
\n2 118363 115 128 0 1
\n3 131167 115 7860 0 1
\n4 137965 115 7922 0 1
\n5 157063 115 156 0 1
\n6 176092 115 51 0 1
\n7 194190 115 50 0 1
\n8 212463 115 7896 0 1
\n9 230983 115 7863 0 1

\n

No lectures. May you help how could I find its lectures?

", "rawMarkdown": "Thank you for answering. No I did not concatenated. I've just explored train df. Let's choose user 115. Head of its df ordered by timestamp:\n timestamp user_id content_id content_type_id answered_correctly\n0 0 115 5692 0 1\n1 56943 115 5716 0 1\n2 118363 115 128 0 1\n3 131167 115 7860 0 1\n4 137965 115 7922 0 1\n5 157063 115 156 0 1\n6 176092 115 51 0 1\n7 194190 115 50 0 1\n8 212463 115 7896 0 1\n9 230983 115 7863 0 1\n\nNo lectures. May you help how could I find its lectures?", "votes": null }, { "id": "1108247", "postDate": "12/10/2020 13:08:46", "content": "

115 user has not watched lectures. user_id==705741139 that i mentioned which has lectures. You can use the below to find lectures

\n
\n

train.query('user_id==115 and content_type_id==1')

\n
", "rawMarkdown": "115 user has not watched lectures. user_id==705741139 that i mentioned which has lectures. You can use the below to find lectures\n>train.query('user_id==115 and content_type_id==1')", "votes": null }, { "id": "1108373", "postDate": "12/10/2020 15:28:43", "content": "
\n

Typically the first several questions a user sees were part of an onboarding diagnostic test where they did not get any feedback.

\n
\n

According to the data description, users are first given some sample questions to answer as a part of the onboarding diagnostic test.

", "rawMarkdown": "> Typically the first several questions a user sees were part of an onboarding diagnostic test where they did not get any feedback.\n\nAccording to the data description, users are first given some sample questions to answer as a part of the onboarding diagnostic test.", "votes": null }, { "id": "1108389", "postDate": "12/10/2020 15:39:12", "content": "

Definitely I can take. Just trying to understand the cause of such behaviour. Anyway thank you for your help!

", "rawMarkdown": "Definitely I can take. Just trying to understand the cause of such behaviour. Anyway thank you for your help!", "votes": null }, { "id": "1108391", "postDate": "12/10/2020 15:39:54", "content": "

Thanks. I had such such idea.

", "rawMarkdown": "Thanks. I had such such idea.", "votes": null } ], "comments": [ { "id": 1108238, "author_name": "watzisname", "author_url": "", "post_date": "12/10/2020 12:52:57", "content": "

That is not the case - I just checked a user, user_id ==705741139.
\nDid you concat your questions and lectures dataframes - in which case you will see lectures at the end.

", "votes": null, "replies": [ { "id": 1108243, "author_name": "xyutech", "author_url": "", "post_date": "12/10/2020 13:02:18", "content": "

Thank you for answering. No I did not concatenated. I've just explored train df. Let's choose user 115. Head of its df ordered by timestamp:
\n timestamp user_id content_id content_type_id answered_correctly
\n0 0 115 5692 0 1
\n1 56943 115 5716 0 1
\n2 118363 115 128 0 1
\n3 131167 115 7860 0 1
\n4 137965 115 7922 0 1
\n5 157063 115 156 0 1
\n6 176092 115 51 0 1
\n7 194190 115 50 0 1
\n8 212463 115 7896 0 1
\n9 230983 115 7863 0 1

\n

No lectures. May you help how could I find its lectures?

", "votes": null, "replies": [] }, { "id": 1108247, "author_name": "watzisname", "author_url": "", "post_date": "12/10/2020 13:08:46", "content": "

115 user has not watched lectures. user_id==705741139 that i mentioned which has lectures. You can use the below to find lectures

\n
\n

train.query('user_id==115 and content_type_id==1')

\n
", "votes": null, "replies": [] }, { "id": 1108389, "author_name": "xyutech", "author_url": "", "post_date": "12/10/2020 15:39:12", "content": "

Definitely I can take. Just trying to understand the cause of such behaviour. Anyway thank you for your help!

", "votes": null, "replies": [] } ] }, { "id": 1108373, "author_name": "manikanthr5", "author_url": "", "post_date": "12/10/2020 15:28:43", "content": "
\n

Typically the first several questions a user sees were part of an onboarding diagnostic test where they did not get any feedback.

\n
\n

According to the data description, users are first given some sample questions to answer as a part of the onboarding diagnostic test.

", "votes": null, "replies": [ { "id": 1108391, "author_name": "xyutech", "author_url": "", "post_date": "12/10/2020 15:39:54", "content": "

Thanks. I had such such idea.

", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "1107985": "Hello to everyone,\nI've started analysing learning paths for randomly selected users and found that for all questions go before lectures? Why it is so? I'd expected lectures first then questions then next cycle.", "1108238": "That is not the case - I just checked a user, user_id ==705741139. \nDid you concat your questions and lectures dataframes - in which case you will see lectures at the end.", "1108243": "Thank you for answering. No I did not concatenated. I've just explored train df. Let's choose user 115. Head of its df ordered by timestamp:\n timestamp user_id content_id content_type_id answered_correctly\n0 0 115 5692 0 1\n1 56943 115 5716 0 1\n2 118363 115 128 0 1\n3 131167 115 7860 0 1\n4 137965 115 7922 0 1\n5 157063 115 156 0 1\n6 176092 115 51 0 1\n7 194190 115 50 0 1\n8 212463 115 7896 0 1\n9 230983 115 7863 0 1\n\nNo lectures. May you help how could I find its lectures?", "1108247": "115 user has not watched lectures. user_id==705741139 that i mentioned which has lectures. You can use the below to find lectures\n>train.query('user_id==115 and content_type_id==1')", "1108373": "> Typically the first several questions a user sees were part of an onboarding diagnostic test where they did not get any feedback.\n\nAccording to the data description, users are first given some sample questions to answer as a part of the onboarding diagnostic test.", "1108389": "Definitely I can take. Just trying to understand the cause of such behaviour. Anyway thank you for your help!", "1108391": "Thanks. I had such such idea." }, "source": "meta" }