{
  "id": 527003,
  "title": "Extended leaderboards (9th August)",
  "url": "/competitions/the-future-crop-challenge/discussion/527003",
  "author_name": "Lily-belle Sweet",
  "post_date": "2024-08-09T13:14:58.042000",
  "votes": 2,
  "comment_count": 0,
  "views": 0,
  "content": "<p>We're not only looking for the best performance in terms of RMSE, but also in terms of how well the models can capture the variability of annual production in breadbasket regions (Iowa for maize, Germany for wheat). Furthermore, as we find that often models overfit spatially but don't manage to get the temporal variability at each gridcell, we also measure the median R2 over all gridcells.</p>\n<p>The number next to your first name or initial indicates which of your submissions this represents - 1 would be the first you submitted, etc.</p>\n<p>The full leaderboard is available in a <a href=\"https://docs.google.com/spreadsheets/d/18dVIa7DPN_EX23ZZs2IdFxs66B_vQRwcpXIFvAU2tC8/edit?usp=sharing\" target=\"_blank\">Google Sheet</a> For the top ten submissions for each metric, see below.</p>\n<p>Should we consider other metrics as well? Interested in your thoughts!</p>\n<p><strong>Don't forget - to enter the competition it's compulsory to share your code. Please make your notebooks public or contact us to share!</strong></p>\n<h3>Median R^2, top ten</h3>\n<table>\n<thead>\n<tr>\n<th>Submission</th>\n<th>Validation Median R2</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>Abdelghani54</td>\n<td>0.1605</td>\n</tr>\n<tr>\n<td>Abdelghani53</td>\n<td>0.1549</td>\n</tr>\n<tr>\n<td>Abdelghani52</td>\n<td>0.1538</td>\n</tr>\n<tr>\n<td>Abdelghani50</td>\n<td>0.1526</td>\n</tr>\n<tr>\n<td>Abdelghani35</td>\n<td>0.1513</td>\n</tr>\n<tr>\n<td>Abdelghani38</td>\n<td>0.1496</td>\n</tr>\n<tr>\n<td>Abdelghani51</td>\n<td>0.1484</td>\n</tr>\n<tr>\n<td>Abdelghani36</td>\n<td>0.147</td>\n</tr>\n<tr>\n<td>Abdelghani37</td>\n<td>0.1468</td>\n</tr>\n<tr>\n<td>Monique26</td>\n<td>0.1402</td>\n</tr>\n</tbody>\n</table>\n<h3>Iowa maize R^2, top ten</h3>\n<table>\n<thead>\n<tr>\n<th>Submission</th>\n<th>Validation Iowa R2</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>Monique17</td>\n<td>0.7599</td>\n</tr>\n<tr>\n<td>AmittKumarrSrivastava5</td>\n<td>0.7597</td>\n</tr>\n<tr>\n<td>Monique16</td>\n<td>0.7025</td>\n</tr>\n<tr>\n<td>Jingye1</td>\n<td>0.6672</td>\n</tr>\n<tr>\n<td>Monique1</td>\n<td>0.6535</td>\n</tr>\n<tr>\n<td>Monique6</td>\n<td>0.6531</td>\n</tr>\n<tr>\n<td>AmittKumarrSrivastava4</td>\n<td>0.6285</td>\n</tr>\n<tr>\n<td>Abdelghani27</td>\n<td>0.617</td>\n</tr>\n<tr>\n<td>Abdelghani25</td>\n<td>0.6169</td>\n</tr>\n<tr>\n<td>Monique10</td>\n<td>0.6148</td>\n</tr>\n</tbody>\n</table>\n<h3>Germany wheat R^2, top ten</h3>\n<table>\n<thead>\n<tr>\n<th>Submission</th>\n<th>Validation Germany R2</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>Johannes9</td>\n<td>0.2706</td>\n</tr>\n<tr>\n<td>Monique10</td>\n<td>0.2401</td>\n</tr>\n<tr>\n<td>K1</td>\n<td>0.235</td>\n</tr>\n<tr>\n<td>K2</td>\n<td>0.235</td>\n</tr>\n<tr>\n<td>Monique12</td>\n<td>0.2064</td>\n</tr>\n<tr>\n<td>Monique17</td>\n<td>0.1983</td>\n</tr>\n<tr>\n<td>AmittKumarrSrivastava5</td>\n<td>0.1903</td>\n</tr>\n<tr>\n<td>Divya1</td>\n<td>0.1801</td>\n</tr>\n<tr>\n<td>Abdelghani17</td>\n<td>0.1719</td>\n</tr>\n<tr>\n<td>JasonTarzan3</td>\n<td>0.1715</td>\n</tr>\n</tbody>\n</table>",
  "messages": [
    {
      "id": 2954196,
      "postDate": "2024-08-09T13:14:58.043Z",
      "content": "<p>We're not only looking for the best performance in terms of RMSE, but also in terms of how well the models can capture the variability of annual production in breadbasket regions (Iowa for maize, Germany for wheat). Furthermore, as we find that often models overfit spatially but don't manage to get the temporal variability at each gridcell, we also measure the median R2 over all gridcells.</p>\n<p>The number next to your first name or initial indicates which of your submissions this represents - 1 would be the first you submitted, etc.</p>\n<p>The full leaderboard is available in a <a href=\"https://docs.google.com/spreadsheets/d/18dVIa7DPN_EX23ZZs2IdFxs66B_vQRwcpXIFvAU2tC8/edit?usp=sharing\" target=\"_blank\">Google Sheet</a> For the top ten submissions for each metric, see below.</p>\n<p>Should we consider other metrics as well? Interested in your thoughts!</p>\n<p><strong>Don't forget - to enter the competition it's compulsory to share your code. Please make your notebooks public or contact us to share!</strong></p>\n<h3>Median R^2, top ten</h3>\n<table>\n<thead>\n<tr>\n<th>Submission</th>\n<th>Validation Median R2</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>Abdelghani54</td>\n<td>0.1605</td>\n</tr>\n<tr>\n<td>Abdelghani53</td>\n<td>0.1549</td>\n</tr>\n<tr>\n<td>Abdelghani52</td>\n<td>0.1538</td>\n</tr>\n<tr>\n<td>Abdelghani50</td>\n<td>0.1526</td>\n</tr>\n<tr>\n<td>Abdelghani35</td>\n<td>0.1513</td>\n</tr>\n<tr>\n<td>Abdelghani38</td>\n<td>0.1496</td>\n</tr>\n<tr>\n<td>Abdelghani51</td>\n<td>0.1484</td>\n</tr>\n<tr>\n<td>Abdelghani36</td>\n<td>0.147</td>\n</tr>\n<tr>\n<td>Abdelghani37</td>\n<td>0.1468</td>\n</tr>\n<tr>\n<td>Monique26</td>\n<td>0.1402</td>\n</tr>\n</tbody>\n</table>\n<h3>Iowa maize R^2, top ten</h3>\n<table>\n<thead>\n<tr>\n<th>Submission</th>\n<th>Validation Iowa R2</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>Monique17</td>\n<td>0.7599</td>\n</tr>\n<tr>\n<td>AmittKumarrSrivastava5</td>\n<td>0.7597</td>\n</tr>\n<tr>\n<td>Monique16</td>\n<td>0.7025</td>\n</tr>\n<tr>\n<td>Jingye1</td>\n<td>0.6672</td>\n</tr>\n<tr>\n<td>Monique1</td>\n<td>0.6535</td>\n</tr>\n<tr>\n<td>Monique6</td>\n<td>0.6531</td>\n</tr>\n<tr>\n<td>AmittKumarrSrivastava4</td>\n<td>0.6285</td>\n</tr>\n<tr>\n<td>Abdelghani27</td>\n<td>0.617</td>\n</tr>\n<tr>\n<td>Abdelghani25</td>\n<td>0.6169</td>\n</tr>\n<tr>\n<td>Monique10</td>\n<td>0.6148</td>\n</tr>\n</tbody>\n</table>\n<h3>Germany wheat R^2, top ten</h3>\n<table>\n<thead>\n<tr>\n<th>Submission</th>\n<th>Validation Germany R2</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>Johannes9</td>\n<td>0.2706</td>\n</tr>\n<tr>\n<td>Monique10</td>\n<td>0.2401</td>\n</tr>\n<tr>\n<td>K1</td>\n<td>0.235</td>\n</tr>\n<tr>\n<td>K2</td>\n<td>0.235</td>\n</tr>\n<tr>\n<td>Monique12</td>\n<td>0.2064</td>\n</tr>\n<tr>\n<td>Monique17</td>\n<td>0.1983</td>\n</tr>\n<tr>\n<td>AmittKumarrSrivastava5</td>\n<td>0.1903</td>\n</tr>\n<tr>\n<td>Divya1</td>\n<td>0.1801</td>\n</tr>\n<tr>\n<td>Abdelghani17</td>\n<td>0.1719</td>\n</tr>\n<tr>\n<td>JasonTarzan3</td>\n<td>0.1715</td>\n</tr>\n</tbody>\n</table>",
      "rawMarkdown": "We're not only looking for the best performance in terms of RMSE, but also in terms of how well the models can capture the variability of annual production in breadbasket regions (Iowa for maize, Germany for wheat). Furthermore, as we find that often models overfit spatially but don't manage to get the temporal variability at each gridcell, we also measure the median R2 over all gridcells.\n\nThe number next to your first name or initial indicates which of your submissions this represents - 1 would be the first you submitted, etc.\n\nThe full leaderboard is available in a [Google Sheet](https://docs.google.com/spreadsheets/d/18dVIa7DPN_EX23ZZs2IdFxs66B_vQRwcpXIFvAU2tC8/edit?usp=sharing) For the top ten submissions for each metric, see below.\n\nShould we consider other metrics as well? Interested in your thoughts!\n\n**Don't forget - to enter the competition it's compulsory to share your code. Please make your notebooks public or contact us to share!**\n\n### Median R^2, top ten\n\nSubmission|Validation Median R2\n---|---\nAbdelghani54|0.1605\nAbdelghani53|0.1549\nAbdelghani52|0.1538\nAbdelghani50|0.1526\nAbdelghani35|0.1513\nAbdelghani38|0.1496\nAbdelghani51|0.1484\nAbdelghani36|0.147\nAbdelghani37|0.1468\nMonique26|0.1402\n\n### Iowa maize R^2, top ten\n\nSubmission|Validation Iowa R2\n---|---\nMonique17|0.7599\nAmittKumarrSrivastava5|0.7597\nMonique16|0.7025\nJingye1|0.6672\nMonique1|0.6535\nMonique6|0.6531\nAmittKumarrSrivastava4|0.6285\nAbdelghani27|0.617\nAbdelghani25|0.6169\nMonique10|0.6148\n\n### Germany wheat R^2, top ten\n\nSubmission|Validation Germany R2\n---|---\nJohannes9|0.2706\nMonique10|0.2401\nK1|0.235\nK2|0.235\nMonique12|0.2064\nMonique17|0.1983\nAmittKumarrSrivastava5|0.1903\nDivya1|0.1801\nAbdelghani17|0.1719\nJasonTarzan3|0.1715",
      "votes": 2
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2954196": "We're not only looking for the best performance in terms of RMSE, but also in terms of how well the models can capture the variability of annual production in breadbasket regions (Iowa for maize, Germany for wheat). Furthermore, as we find that often models overfit spatially but don't manage to get the temporal variability at each gridcell, we also measure the median R2 over all gridcells.\n\nThe number next to your first name or initial indicates which of your submissions this represents - 1 would be the first you submitted, etc.\n\nThe full leaderboard is available in a [Google Sheet](https://docs.google.com/spreadsheets/d/18dVIa7DPN_EX23ZZs2IdFxs66B_vQRwcpXIFvAU2tC8/edit?usp=sharing) For the top ten submissions for each metric, see below.\n\nShould we consider other metrics as well? Interested in your thoughts!\n\n**Don't forget - to enter the competition it's compulsory to share your code. Please make your notebooks public or contact us to share!**\n\n### Median R^2, top ten\n\nSubmission|Validation Median R2\n---|---\nAbdelghani54|0.1605\nAbdelghani53|0.1549\nAbdelghani52|0.1538\nAbdelghani50|0.1526\nAbdelghani35|0.1513\nAbdelghani38|0.1496\nAbdelghani51|0.1484\nAbdelghani36|0.147\nAbdelghani37|0.1468\nMonique26|0.1402\n\n### Iowa maize R^2, top ten\n\nSubmission|Validation Iowa R2\n---|---\nMonique17|0.7599\nAmittKumarrSrivastava5|0.7597\nMonique16|0.7025\nJingye1|0.6672\nMonique1|0.6535\nMonique6|0.6531\nAmittKumarrSrivastava4|0.6285\nAbdelghani27|0.617\nAbdelghani25|0.6169\nMonique10|0.6148\n\n### Germany wheat R^2, top ten\n\nSubmission|Validation Germany R2\n---|---\nJohannes9|0.2706\nMonique10|0.2401\nK1|0.235\nK2|0.235\nMonique12|0.2064\nMonique17|0.1983\nAmittKumarrSrivastava5|0.1903\nDivya1|0.1801\nAbdelghani17|0.1719\nJasonTarzan3|0.1715"
  }
}