{
  "id": 251368,
  "title": "Submission Scoring Error",
  "url": "/competitions/mlb-player-digital-engagement-forecasting/discussion/251368",
  "author_name": "",
  "post_date": "2021-07-07T03:15:07.257672200Z",
  "votes": 4,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I have been working on this competition for several weeks now. However every submission I made gave me a submission scoring error, even-though everything seems to be correct. I have no clue why this is not working could someone please help?</p>\n<p>Thanks in advance<br>\nThis is the <a href=\"https://www.kaggle.com/julius1r/my-submission\" target=\"_blank\">notebook</a></p>",
  "messages": [
    {
      "id": "1379006",
      "postDate": "07/07/2021 03:15:07",
      "content": "<p>I have been working on this competition for several weeks now. However every submission I made gave me a submission scoring error, even-though everything seems to be correct. I have no clue why this is not working could someone please help?</p>\n<p>Thanks in advance<br>\nThis is the <a href=\"https://www.kaggle.com/julius1r/my-submission\" target=\"_blank\">notebook</a></p>",
      "rawMarkdown": "I have been working on this competition for several weeks now. However every submission I made gave me a submission scoring error, even-though everything seems to be correct. I have no clue why this is not working could someone please help?\n\nThanks in advance\nThis is the [notebook](https://www.kaggle.com/julius1r/my-submission)",
      "votes": null
    },
    {
      "id": "1379046",
      "postDate": "07/07/2021 04:26:15",
      "content": "<p>Honestly speaking, I'm facing such problem too! And i think problem is arising in that part</p>\n<pre><code># Dealing with missing values\n    if test_df['rosters'].iloc[0] == test_df['rosters'].iloc[0]:\n        test_rosters = pd.DataFrame(eval(test_df['rosters'].iloc[0]))\n    else:\n        test_rosters = pd.DataFrame({'playerId': sample_prediction_df['playerId']})\n        for col in rosters.columns:\n            if col == 'playerId': continue\n            test_rosters[col] = np.nan\n</code></pre>\n<p>In a submission I completely avoided this part. That means I tried making prediction without rosters data and utilised only players stats (mean, std, max, min) and the submission was succeeded. <br>\nHere is the notebook where I investigated the problem: <br>\n<a href=\"https://www.kaggle.com/pritommojumder/mlb-forecasting-model-xst-inference?scriptVersionId=67641005\" target=\"_blank\">https://www.kaggle.com/pritommojumder/mlb-forecasting-model-xst-inference?scriptVersionId=67641005</a></p>\n<p>Check <strong>Version 9 and Version 10</strong> and you will clearly see the difference. I guess we both need a solution for this.</p>",
      "rawMarkdown": "Honestly speaking, I'm facing such problem too! And i think problem is arising in that part\n```\n# Dealing with missing values\n    if test_df['rosters'].iloc[0] == test_df['rosters'].iloc[0]:\n        test_rosters = pd.DataFrame(eval(test_df['rosters'].iloc[0]))\n    else:\n        test_rosters = pd.DataFrame({'playerId': sample_prediction_df['playerId']})\n        for col in rosters.columns:\n            if col == 'playerId': continue\n            test_rosters[col] = np.nan\n```\nIn a submission I completely avoided this part. That means I tried making prediction without rosters data and utilised only players stats (mean, std, max, min) and the submission was succeeded. \nHere is the notebook where I investigated the problem: \nhttps://www.kaggle.com/pritommojumder/mlb-forecasting-model-xst-inference?scriptVersionId=67641005\n\nCheck **Version 9 and Version 10** and you will clearly see the difference. I guess we both need a solution for this.",
      "votes": null
    },
    {
      "id": "1379443",
      "postDate": "07/07/2021 11:27:57",
      "content": "<p>I have the same problem with rosters data</p>",
      "rawMarkdown": "I have the same problem with rosters data",
      "votes": null
    },
    {
      "id": "1380611",
      "postDate": "07/08/2021 07:42:31",
      "content": "<p>Thank you for sharing. It's interesting in the difference with Version 9 and Version 10.  Do you have any idea about that error occur if you just add above rosters scripts you mentioned. Everyone use this rosters scripts without error. </p>",
      "rawMarkdown": "Thank you for sharing. It's interesting in the difference with Version 9 and Version 10.  Do you have any idea about that error occur if you just add above rosters scripts you mentioned. Everyone use this rosters scripts without error.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1379046,
      "author_name": "pritommojumder",
      "author_url": "",
      "post_date": "07/07/2021 04:26:15",
      "content": "<p>Honestly speaking, I'm facing such problem too! And i think problem is arising in that part</p>\n<pre><code># Dealing with missing values\n    if test_df['rosters'].iloc[0] == test_df['rosters'].iloc[0]:\n        test_rosters = pd.DataFrame(eval(test_df['rosters'].iloc[0]))\n    else:\n        test_rosters = pd.DataFrame({'playerId': sample_prediction_df['playerId']})\n        for col in rosters.columns:\n            if col == 'playerId': continue\n            test_rosters[col] = np.nan\n</code></pre>\n<p>In a submission I completely avoided this part. That means I tried making prediction without rosters data and utilised only players stats (mean, std, max, min) and the submission was succeeded. <br>\nHere is the notebook where I investigated the problem: <br>\n<a href=\"https://www.kaggle.com/pritommojumder/mlb-forecasting-model-xst-inference?scriptVersionId=67641005\" target=\"_blank\">https://www.kaggle.com/pritommojumder/mlb-forecasting-model-xst-inference?scriptVersionId=67641005</a></p>\n<p>Check <strong>Version 9 and Version 10</strong> and you will clearly see the difference. I guess we both need a solution for this.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1380611,
          "author_name": "rautaki0127",
          "author_url": "",
          "post_date": "07/08/2021 07:42:31",
          "content": "<p>Thank you for sharing. It's interesting in the difference with Version 9 and Version 10.  Do you have any idea about that error occur if you just add above rosters scripts you mentioned. Everyone use this rosters scripts without error. </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1379443,
      "author_name": "enric1296",
      "author_url": "",
      "post_date": "07/07/2021 11:27:57",
      "content": "<p>I have the same problem with rosters data</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1379006": "I have been working on this competition for several weeks now. However every submission I made gave me a submission scoring error, even-though everything seems to be correct. I have no clue why this is not working could someone please help?\n\nThanks in advance\nThis is the [notebook](https://www.kaggle.com/julius1r/my-submission)",
    "1379046": "Honestly speaking, I'm facing such problem too! And i think problem is arising in that part\n```\n# Dealing with missing values\n    if test_df['rosters'].iloc[0] == test_df['rosters'].iloc[0]:\n        test_rosters = pd.DataFrame(eval(test_df['rosters'].iloc[0]))\n    else:\n        test_rosters = pd.DataFrame({'playerId': sample_prediction_df['playerId']})\n        for col in rosters.columns:\n            if col == 'playerId': continue\n            test_rosters[col] = np.nan\n```\nIn a submission I completely avoided this part. That means I tried making prediction without rosters data and utilised only players stats (mean, std, max, min) and the submission was succeeded. \nHere is the notebook where I investigated the problem: \nhttps://www.kaggle.com/pritommojumder/mlb-forecasting-model-xst-inference?scriptVersionId=67641005\n\nCheck **Version 9 and Version 10** and you will clearly see the difference. I guess we both need a solution for this.",
    "1379443": "I have the same problem with rosters data",
    "1380611": "Thank you for sharing. It's interesting in the difference with Version 9 and Version 10.  Do you have any idea about that error occur if you just add above rosters scripts you mentioned. Everyone use this rosters scripts without error."
  },
  "source": "meta"
}