{ "id": 399165, "title": "Where is the sample notebook referred to in the `Data` tab?", "url": "/competitions/predict-student-performance-from-game-play/discussion/399165", "author_name": "calvdee", "post_date": "2023-04-02T20:03:12.058000", "votes": 2, "comment_count": 7, "views": 0, "content": "

In the Data tab, it says

\n
\n

use the sample notebook to iterate over the test data

\n
\n

Where do I find this sample notebook?

", "messages": [ { "id": 2206766, "postDate": "2023-04-02T20:20:48.387Z", "content": "

I also didn't find mentioned notebook. But I guess Infer test data part of this notebook would work for you.

", "rawMarkdown": "I also didn't find mentioned notebook. But I guess Infer test data part of [this notebook](https://www.kaggle.com/code/cdeotte/xgboost-baseline-0-680) would work for you.", "votes": 1, "replies": [ { "id": 2206776, "postDate": "2023-04-02T20:38:45.013Z", "content": "

Thanks! Will check it out after doing some of my own EDA :)

", "rawMarkdown": "Thanks! Will check it out after doing some of my own EDA :)" }, { "id": 2206803, "postDate": "2023-04-02T21:28:17.700Z", "content": "

I don't understand what this does:

\n
# IMPORT KAGGLE API\nimport jo_wilder\nenv = jo_wilder.make_env()\niter_test = env.iter_test()\n
\n

The __init__.py file just contains:

\n
from .competition import make_env\n\n__all__ = ['make_env']\n
\n

There is no competition module.

\n

Do you know how I should be running this?

", "rawMarkdown": "I don't understand what this does:\n\n```\n# IMPORT KAGGLE API\nimport jo_wilder\nenv = jo_wilder.make_env()\niter_test = env.iter_test()\n```\n\nThe `__init__.py` file just contains:\n\n```\nfrom .competition import make_env\n\n__all__ = ['make_env']\n```\n\nThere is no `competition` module. \n\nDo you know how I should be running this?", "replies": [ { "id": 2207089, "postDate": "2023-04-03T05:30:43.550Z", "content": "

It connects to Kaggle API, you should run it as-is. No need to check init.py. Once you run it, next code to run is:

\n
 (test, sample_submission)  iter_test:\n    sample_submission[] = [(label.split()[][:])  label  sample_submission[]]\n\n    \n    df = feature_engineer(test)\n\n    \n    grp = test.level_group.values[]\n    a,b = limits[grp]\n     t  (a,b):\n        clf = models[]\n        p = clf.predict_proba(df[FEATURES].astype())[:,]\n        mask = sample_submission.question == t    \n        sample_submission.loc[mask, ] = (p > best_threshold).astype() \n    env.predict(sample_submission[[, ]])\n
\n

Basically, this code gives you access to one level group at a time from test data and you can give predictions only here.

", "rawMarkdown": "It connects to Kaggle API, you should run it as-is. No need to check __init__.py. Once you run it, next code to run is:\n\n\n```python\nfor (test, sample_submission) in iter_test:\n sample_submission['question'] = [int(label.split('_')[1][1:]) for label in sample_submission['session_id']]\n\n # Your function to transform test data\n df = feature_engineer(test)\n \n # INFER TEST DATA\n grp = test.level_group.values[0]\n a,b = limits[grp]\n for t in range(a,b):\n clf = models[f'{grp}_{t}']\n p = clf.predict_proba(df[FEATURES].astype('float32'))[:,1]\n mask = sample_submission.question == t \n sample_submission.loc[mask, 'correct'] = (p > best_threshold).astype('int') \n env.predict(sample_submission[['session_id', 'correct']])\n```\n\nBasically, this code gives you access to one level group at a time from test data and you can give predictions only here.\n\n" }, { "id": 2209814, "postDate": "2023-04-04T23:52:34.860Z", "rawMarkdown": "", "isDeleted": true } ] } ] }, { "id": 2206755, "postDate": "2023-04-02T20:03:12.060Z", "content": "

In the Data tab, it says

\n
\n

use the sample notebook to iterate over the test data

\n
\n

Where do I find this sample notebook?

", "rawMarkdown": "In the `Data` tab, it says \n\n> use the sample notebook to iterate over the test data\n\nWhere do I find this sample notebook?", "votes": 1 }, { "id": 2208523, "postDate": "2023-04-04T06:03:31.337Z", "content": "

The official sample notebook is here, but it doesn't work anymore because Kaggle updated the API and now the API returns for (test, sample_submission) in iter_test: instead of for (sample_submission, test) in iter_test: so Elshan Kazim suggestion of using my notebook is a good one because my notebook works with the new API

", "rawMarkdown": "The official sample notebook is [here][1], but it doesn't work anymore because Kaggle updated the API and now the API returns `for (test, sample_submission) in iter_test:` instead of `for (sample_submission, test) in iter_test:` so Elshan Kazim suggestion of using my notebook is a good one because my notebook works with the new API\n\n[1]: https://www.kaggle.com/code/philculliton/basic-submission-demo", "replies": [ { "id": 2209816, "postDate": "2023-04-04T23:56:18.537Z", "content": "

Thanks for putting the notebook together. I haven't participated in a comp in a while…Sounds like the only way to make predictions then is using Kaggle hosted notebooks (vs doing it say locally or in another hosted environment)?

", "rawMarkdown": "Thanks for putting the notebook together. I haven't participated in a comp in a while...Sounds like the only way to make predictions then is using Kaggle hosted notebooks (vs doing it say locally or in another hosted environment)?" } ] } ], "comments": [ { "id": 2206766, "author_name": "Elshan Kazim", "author_url": "", "post_date": "2023-04-02T20:20:48.387000", "content": "

I also didn't find mentioned notebook. But I guess Infer test data part of this notebook would work for you.

", "votes": 1, "replies": [ { "id": 2206776, "author_name": "calvdee", "author_url": "", "post_date": "2023-04-02T20:38:45.013000", "content": "

Thanks! Will check it out after doing some of my own EDA :)

", "votes": 0, "replies": [] }, { "id": 2206803, "author_name": "calvdee", "author_url": "", "post_date": "2023-04-02T21:28:17.700000", "content": "

I don't understand what this does:

\n
# IMPORT KAGGLE API\nimport jo_wilder\nenv = jo_wilder.make_env()\niter_test = env.iter_test()\n
\n

The __init__.py file just contains:

\n
from .competition import make_env\n\n__all__ = ['make_env']\n
\n

There is no competition module.

\n

Do you know how I should be running this?

", "votes": 0, "replies": [ { "id": 2207089, "author_name": "Elshan Kazim", "author_url": "", "post_date": "2023-04-03T05:30:43.550000", "content": "

It connects to Kaggle API, you should run it as-is. No need to check init.py. Once you run it, next code to run is:

\n
 (test, sample_submission)  iter_test:\n    sample_submission[] = [(label.split()[][:])  label  sample_submission[]]\n\n    \n    df = feature_engineer(test)\n\n    \n    grp = test.level_group.values[]\n    a,b = limits[grp]\n     t  (a,b):\n        clf = models[]\n        p = clf.predict_proba(df[FEATURES].astype())[:,]\n        mask = sample_submission.question == t    \n        sample_submission.loc[mask, ] = (p > best_threshold).astype() \n    env.predict(sample_submission[[, ]])\n
\n

Basically, this code gives you access to one level group at a time from test data and you can give predictions only here.

", "votes": 0, "replies": [] }, { "id": 2209814, "author_name": "", "author_url": "", "post_date": "2023-04-04T23:52:34.860000", "content": "", "votes": 0, "replies": [] } ] } ] }, { "id": 2208523, "author_name": "Chris Deotte", "author_url": "", "post_date": "2023-04-04T06:03:31.337000", "content": "

The official sample notebook is here, but it doesn't work anymore because Kaggle updated the API and now the API returns for (test, sample_submission) in iter_test: instead of for (sample_submission, test) in iter_test: so Elshan Kazim suggestion of using my notebook is a good one because my notebook works with the new API

", "votes": 0, "replies": [ { "id": 2209816, "author_name": "calvdee", "author_url": "", "post_date": "2023-04-04T23:56:18.537000", "content": "

Thanks for putting the notebook together. I haven't participated in a comp in a while…Sounds like the only way to make predictions then is using Kaggle hosted notebooks (vs doing it say locally or in another hosted environment)?

", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "2206766": "I also didn't find mentioned notebook. But I guess Infer test data part of [this notebook](https://www.kaggle.com/code/cdeotte/xgboost-baseline-0-680) would work for you.", "2206755": "In the `Data` tab, it says \n\n> use the sample notebook to iterate over the test data\n\nWhere do I find this sample notebook?", "2208523": "The official sample notebook is [here][1], but it doesn't work anymore because Kaggle updated the API and now the API returns `for (test, sample_submission) in iter_test:` instead of `for (sample_submission, test) in iter_test:` so Elshan Kazim suggestion of using my notebook is a good one because my notebook works with the new API\n\n[1]: https://www.kaggle.com/code/philculliton/basic-submission-demo" } }