{
  "id": 11087,
  "title": "*URGENT* - ids in Test and SampleSubmission not matching ",
  "url": "/competitions/avazu-ctr-prediction/discussion/11087",
  "author_name": "",
  "post_date": "2014-11-29T16:37:35.960Z",
  "votes": -2,
  "comment_count": 4,
  "views": 1105,
  "content": "<p>Hi Everyone,</p>\n<p>&nbsp; I've downloaded the test dataset on 6th Nov and the sampleSubmission today. I tried to match&nbsp;the ids from the test dataset and sampleSubmission but not a single id is matching between them. I'm not sure if new test dataset was uploaded in kaggle after 6th Nov as I haven't followed the forum regularly after 6th. Apologies for that!!</p>\n\n<p>But did anyone face the same problem? Can anyone tell where to find the latest test dataset or the sampleSubmission? &nbsp;Below is the R code which I used to do the matching: &lt;data is a subfolder&nbsp;in my working directory&gt;</p>\n<p>tst &lt;- read.csv(&quot;data/test.csv&quot;,header = T,sep = &quot;,&quot;);</p>\n<p>sub &lt;- read.csv(file = &quot;data/sampleSubmission.csv&quot;, header = T, sep = &quot;,&quot;);</p>\n<p>x&lt;-intersect(x = sub[,&quot;id&quot;],y=tst[,&quot;id&quot;]);</p>\n<p>x</p>\n<p>and I get</p>\n<p>numeric(0), which is an empty.</p>\n<p>Appreciate if someone can throw some light on this issue!!</p>\n<p>Thanks,</p>\n<p>Avishek</p>\n<p>&nbsp;&nbsp;</p>",
  "messages": [
    {
      "id": "59227",
      "postDate": "11/29/2014 16:37:35",
      "content": "<p>Hi Everyone,</p>\n<p>&nbsp; I've downloaded the test dataset on 6th Nov and the sampleSubmission today. I tried to match&nbsp;the ids from the test dataset and sampleSubmission but not a single id is matching between them. I'm not sure if new test dataset was uploaded in kaggle after 6th Nov as I haven't followed the forum regularly after 6th. Apologies for that!!</p>\n\n<p>But did anyone face the same problem? Can anyone tell where to find the latest test dataset or the sampleSubmission? &nbsp;Below is the R code which I used to do the matching: &lt;data is a subfolder&nbsp;in my working directory&gt;</p>\n<p>tst &lt;- read.csv(&quot;data/test.csv&quot;,header = T,sep = &quot;,&quot;);</p>\n<p>sub &lt;- read.csv(file = &quot;data/sampleSubmission.csv&quot;, header = T, sep = &quot;,&quot;);</p>\n<p>x&lt;-intersect(x = sub[,&quot;id&quot;],y=tst[,&quot;id&quot;]);</p>\n<p>x</p>\n<p>and I get</p>\n<p>numeric(0), which is an empty.</p>\n<p>Appreciate if someone can throw some light on this issue!!</p>\n<p>Thanks,</p>\n<p>Avishek</p>\n<p>&nbsp;&nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "59229",
      "postDate": "11/29/2014 17:20:42",
      "content": "<p>ids should not match, those are the new items you want to make predictions on...</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "59230",
      "postDate": "11/29/2014 17:24:04",
      "content": "<p>ids sud definitely match between test.csv data and sampleSubmission.csv. I wasn't talking about training data set</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "59231",
      "postDate": "11/29/2014 17:24:26",
      "content": "<p>There's a new dataset. You need to download the new lot of data from the Data page</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "59232",
      "postDate": "11/29/2014 17:31:15",
      "content": "<p>Yeah, found that out.&nbsp;Thanks M for confirming. Cheers!!</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 59229,
      "author_name": "tanopereira",
      "author_url": "",
      "post_date": "11/29/2014 17:20:42",
      "content": "<p>ids should not match, those are the new items you want to make predictions on...</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 59230,
      "author_name": "avishek",
      "author_url": "",
      "post_date": "11/29/2014 17:24:04",
      "content": "<p>ids sud definitely match between test.csv data and sampleSubmission.csv. I wasn't talking about training data set</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 59231,
      "author_name": "domcastro",
      "author_url": "",
      "post_date": "11/29/2014 17:24:26",
      "content": "<p>There's a new dataset. You need to download the new lot of data from the Data page</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 59232,
      "author_name": "avishek",
      "author_url": "",
      "post_date": "11/29/2014 17:31:15",
      "content": "<p>Yeah, found that out.&nbsp;Thanks M for confirming. Cheers!!</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "59227": "",
    "59229": "",
    "59230": "",
    "59231": "",
    "59232": ""
  },
  "source": "meta"
}