{
  "id": 80806,
  "title": "Different methods for iterating through test data",
  "url": "/competitions/LANL-Earthquake-Prediction/discussion/80806",
  "author_name": "",
  "post_date": "2019-02-16T18:32:53.507401800Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi everyone,</p>\n\n<p>I wanted to start a discussion for people to share the different methods they have used to iterate through the test data, and the methods to generate a submission file. I believe this would be very helpful for everyone in the competition since there are 2,624 files we need to run our model through. I'm excited to see what you guys have come up with.</p>\n\n<p>Thanks,\nJesus</p>",
  "messages": [
    {
      "id": "472838",
      "postDate": "02/16/2019 18:32:53",
      "content": "<p>Hi everyone,</p>\n\n<p>I wanted to start a discussion for people to share the different methods they have used to iterate through the test data, and the methods to generate a submission file. I believe this would be very helpful for everyone in the competition since there are 2,624 files we need to run our model through. I'm excited to see what you guys have come up with.</p>\n\n<p>Thanks,\nJesus</p>",
      "rawMarkdown": "Hi everyone,\n\nI wanted to start a discussion for people to share the different methods they have used to iterate through the test data, and the methods to generate a submission file. I believe this would be very helpful for everyone in the competition since there are 2,624 files we need to run our model through. I'm excited to see what you guys have come up with.\n\nThanks,\nJesus",
      "votes": null
    },
    {
      "id": "473168",
      "postDate": "02/17/2019 13:15:58",
      "content": "<p>Read the sample submission file, use it to create an empty test_X array/DataFrame and then fill in the values. Use test_X for prediction.</p>\n\n<p><code>submission = pd.read_csv('../input/sample_submission.csv', index_col='seg_id')\ntest_X = np.zeros(shape=(submission.shape[0],train_x.shape[1]]),dtype=)\nfor seg_id in tqdm(range(submission.shape[0])):\n    seg=pd.read_csv('../input/test/'+str(submission.index[seg_id])+'.csv', dtype={'acoustic_data': np.int16, 'time_to_failure': np.float32})\n    test_X = create_features(seg_id,seg,test_X)\nmodel.predict(test_X)</code></p>",
      "rawMarkdown": "Read the sample submission file, use it to create an empty test_X array/DataFrame and then fill in the values. Use test_X for prediction.\n\n\n`submission = pd.read_csv('../input/sample_submission.csv', index_col='seg_id')\ntest_X = np.zeros(shape=(submission.shape[0],train_x.shape[1]]),dtype=)\nfor seg_id in tqdm(range(submission.shape[0])):\n    seg=pd.read_csv('../input/test/'+str(submission.index[seg_id])+'.csv', dtype={'acoustic_data': np.int16, 'time_to_failure': np.float32})\n    test_X = create_features(seg_id,seg,test_X)\nmodel.predict(test_X)`",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 473168,
      "author_name": "abdurrafae",
      "author_url": "",
      "post_date": "02/17/2019 13:15:58",
      "content": "<p>Read the sample submission file, use it to create an empty test_X array/DataFrame and then fill in the values. Use test_X for prediction.</p>\n\n<p><code>submission = pd.read_csv('../input/sample_submission.csv', index_col='seg_id')\ntest_X = np.zeros(shape=(submission.shape[0],train_x.shape[1]]),dtype=)\nfor seg_id in tqdm(range(submission.shape[0])):\n    seg=pd.read_csv('../input/test/'+str(submission.index[seg_id])+'.csv', dtype={'acoustic_data': np.int16, 'time_to_failure': np.float32})\n    test_X = create_features(seg_id,seg,test_X)\nmodel.predict(test_X)</code></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "472838": "Hi everyone,\n\nI wanted to start a discussion for people to share the different methods they have used to iterate through the test data, and the methods to generate a submission file. I believe this would be very helpful for everyone in the competition since there are 2,624 files we need to run our model through. I'm excited to see what you guys have come up with.\n\nThanks,\nJesus",
    "473168": "Read the sample submission file, use it to create an empty test_X array/DataFrame and then fill in the values. Use test_X for prediction.\n\n\n`submission = pd.read_csv('../input/sample_submission.csv', index_col='seg_id')\ntest_X = np.zeros(shape=(submission.shape[0],train_x.shape[1]]),dtype=)\nfor seg_id in tqdm(range(submission.shape[0])):\n    seg=pd.read_csv('../input/test/'+str(submission.index[seg_id])+'.csv', dtype={'acoustic_data': np.int16, 'time_to_failure': np.float32})\n    test_X = create_features(seg_id,seg,test_X)\nmodel.predict(test_X)`"
  },
  "source": "meta"
}