# GEMS Prize Challenge **URL:** https://community.drivendata.org/c/gems-prize-challenge/111.md [Latest](https://community.drivendata.org/latest.md) · [Categories](https://community.drivendata.org/categories.md) --- ## [About the GEMS Prize Challenge category](https://community.drivendata.org/t/about-the-gems-prize-challenge-category/11499)
**Author:** [@chrisk-dd](https://community.drivendata.org/u/chrisk-dd)\ **Replies:** 1\ **Last updated:** [September 10, 2026, 7:58pm UTC](https://community.drivendata.org/t/about-the-gems-prize-challenge-category/11499 "2026-09-10T19:58:21Z")
This category is for the Geologic Enhanced Mapping System (GEMS) Prize Challenge. Please keep all posts here specific to this challenge. --- ## [Submission amount question](https://community.drivendata.org/t/submission-amount-question/11553)
**Author:** [@DARD](https://community.drivendata.org/u/DARD)\ **Replies:** 0\ **Last updated:** [October 4, 2026, 1:41pm UTC](https://community.drivendata.org/t/submission-amount-question/11553 "2026-10-04T13:41:47Z")
Hello GEMS Prize organizers, I’m an individual competitor on DrivenData (username: DARD) and I’d like to clarify the weekly submission limit. The Official Rules state that competitors may submit up to three submissions… --- ## [Leaderboard aggregation: pooled over public-test pixels, or mean of per-chunk scores?](https://community.drivendata.org/t/leaderboard-aggregation-pooled-over-public-test-pixels-or-mean-of-per-chunk-scores/11550)
**Author:** [@joeyfezster](https://community.drivendata.org/u/joeyfezster)\ **Replies:** 1\ **Last updated:** [October 1, 2026, 4:11pm UTC](https://community.drivendata.org/t/leaderboard-aggregation-pooled-over-public-test-pixels-or-mean-of-per-chunk-scores/11550 "2026-10-01T16:11:57Z")
Hi DrivenData team, and thanks again for the earlier clarifications on known-fault masking and the 300 m rule, they were very helpful. One more question on how the distance-weighted Tversky score is aggregated: Is the … --- ## [Using a teammate's geological interpretation as training labels](https://community.drivendata.org/t/using-a-teammates-geological-interpretation-as-training-labels/11543)
**Author:** [@exposed](https://community.drivendata.org/u/exposed)\ **Replies:** 1\ **Last updated:** [October 1, 2026, 4:07pm UTC](https://community.drivendata.org/t/using-a-teammates-geological-interpretation-as-training-labels/11543 "2026-10-01T16:07:43Z")
Hi, the rules encourage teams that combine geoscience and data science. If a geologist on our team reviews our model’s candidate faults on the public lidar (accept or reject each one, and occasionally trace a fault), may… --- ## [Looking for a geologist teammate (lidar fault mapping, Basin and Range) to join a working ML pipeline](https://community.drivendata.org/t/looking-for-a-geologist-teammate-lidar-fault-mapping-basin-and-range-to-join-a-working-ml-pipeline/11531)
**Author:** [@exposed](https://community.drivendata.org/u/exposed)\ **Replies:** 3\ **Last updated:** [September 29, 2026, 4:39am UTC](https://community.drivendata.org/t/looking-for-a-geologist-teammate-lidar-fault-mapping-basin-and-range-to-join-a-working-ml-pipeline/11531 "2026-09-29T04:39:53Z")
Hi all, I’m competing solo and looking for one teammate with a geoscience background, ideally someone who has mapped Quaternary faults or scarps on lidar in Nevada or the Walker Lane. Where I am: a working end-to-end p… --- ## [Team member eligibility (competition homepage vs official rules)](https://community.drivendata.org/t/team-member-eligibility-competition-homepage-vs-official-rules/11540)
**Author:** [@dmitry\_v](https://community.drivendata.org/u/dmitry_v)\ **Replies:** 1\ **Last updated:** [September 29, 2026, 2:47am UTC](https://community.drivendata.org/t/team-member-eligibility-competition-homepage-vs-official-rules/11540 "2026-09-29T02:47:16Z")
Hi organizers, The competition homepage states: “Teams must have a captain who is a U.S. citizen or permanent resident.” The Official Rules (Section 1.3) add a requirement for the other team members: “Individuals compe… --- ## [How were the new test faults identified? (data sources and fault types)](https://community.drivendata.org/t/how-were-the-new-test-faults-identified-data-sources-and-fault-types/11527)
**Author:** [@jgaines](https://community.drivendata.org/u/jgaines)\ **Replies:** 9\ **Last updated:** [September 23, 2026, 11:17pm UTC](https://community.drivendata.org/t/how-were-the-new-test-faults-identified-data-sources-and-fault-types/11527 "2026-09-23T23:17:17Z")
Hello. Could we please ask three questions about data sources and fault types involved in identifying the new test faults? We’d like to have better insight into how they were identified, so we can choose appropriate inpu… --- ## [Paid for external data license](https://community.drivendata.org/t/paid-for-external-data-license/11528)
**Author:** [@fishnchips](https://community.drivendata.org/u/fishnchips)\ **Replies:** 1\ **Last updated:** [September 23, 2026, 5:01pm UTC](https://community.drivendata.org/t/paid-for-external-data-license/11528 "2026-09-23T17:01:27Z")
Can we utilise paid for external data, eg if a subscription is required, but if the competition hosts can also access the data - provided they have purchased a license would that solution be allowed? --- ## [Where do you draw the line?](https://community.drivendata.org/t/where-do-you-draw-the-line/11536)
**Author:** [@moongrega](https://community.drivendata.org/u/moongrega)\ **Replies:** 1\ **Last updated:** [September 23, 2026, 7:39am UTC](https://community.drivendata.org/t/where-do-you-draw-the-line/11536 "2026-09-23T07:39:45Z")
I’m still curious where you draw the line between a correction to an existing fault and a new fault. Like, if a mapped USGS/INGENIOUS fault appears to continue beyond its existing endpoint, can that previously unmapped … --- ## [Why does the \`Training fault labels\` file in the data tab have a single band, while the labels in the \`reference solution\` repo have 19 bands?](https://community.drivendata.org/t/why-does-the-training-fault-labels-file-in-the-data-tab-have-a-single-band-while-the-labels-in-the-reference-solution-repo-have-19-bands/11529)
**Author:** [@DLegorreta](https://community.drivendata.org/u/DLegorreta)\ **Replies:** 1\ **Last updated:** [September 23, 2026, 7:36am UTC](https://community.drivendata.org/t/why-does-the-training-fault-labels-file-in-the-data-tab-have-a-single-band-while-the-labels-in-the-reference-solution-repo-have-19-bands/11529 "2026-09-23T07:36:53Z")
here the link of the repo: gems-prize-reference-solution/unet-mc-cv-reference-solution.ipynb at main · drivendataorg/gems-prize-reference-solution · GitHub --- ## [Scoring clarification: are known USGS/INGENIOUS faults masked when scoring, and are they in the Final Round label set?](https://community.drivendata.org/t/scoring-clarification-are-known-usgs-ingenious-faults-masked-when-scoring-and-are-they-in-the-final-round-label-set/11516)
**Author:** [@exposed](https://community.drivendata.org/u/exposed)\ **Replies:** 3\ **Last updated:** [September 21, 2026, 7:25am UTC](https://community.drivendata.org/t/scoring-clarification-are-known-usgs-ingenious-faults-masked-when-scoring-and-are-they-in-the-final-round-label-set/11516 "2026-09-21T07:25:04Z")
Hi organizers, Two questions about how submissions are scored, since they change the optimal way to build a submission: The problem description says the test set consists of newly identified faults that are not in th… --- ## [Institutional Limit](https://community.drivendata.org/t/institutional-limit/11526)
**Author:** [@GGKea](https://community.drivendata.org/u/GGKea)\ **Replies:** 0\ **Last updated:** [September 18, 2026, 10:23pm UTC](https://community.drivendata.org/t/institutional-limit/11526 "2026-09-18T22:23:04Z")
Regarding the rules that, in part, state, “Each participating entity (team, organization, or individual prize competitor not on a team) is allowed to have one final submission; individuals participating on a team will no… --- ## [Weekly Submissions](https://community.drivendata.org/t/weekly-submissions/11524)
**Author:** [@TommyCheeze](https://community.drivendata.org/u/TommyCheeze)\ **Replies:** 1\ **Last updated:** [September 17, 2026, 1:48pm UTC](https://community.drivendata.org/t/weekly-submissions/11524 "2026-09-17T13:48:00Z")
What day and time does the three-submissions-per-week allowance reset, and is it a calendar week or a rolling seven-day window?