# PETs Prize Challenge **URL:** https://community.drivendata.org/c/pets-prize-federated-learning/88.md [Latest](https://community.drivendata.org/latest.md) · [Categories](https://community.drivendata.org/categories.md) --- ## [About the PETs Prize Challenge category](https://community.drivendata.org/t/about-the-pets-prize-challenge-category/7872)
**Author:** [@jayqi](https://community.drivendata.org/u/jayqi)\ **Replies:** 3\ **Last updated:** [July 29, 2022, 11:55am UTC](https://community.drivendata.org/t/about-the-pets-prize-challenge-category/7872 "2022-07-29T11:55:05Z")
This category is for posts about the PETs Prize Challenge: Advancing Privacy-Preserving Federated Learning competition. Please keep all posts here specific to the competition. You can also join the cross-U.S.–U.K. publ… --- ## [About PETs Prize Challenge next year](https://community.drivendata.org/t/about-pets-prize-challenge-next-year/9874)
**Author:** [@kubao](https://community.drivendata.org/u/kubao)\ **Replies:** 1\ **Last updated:** [January 3, 2024, 3:07pm UTC](https://community.drivendata.org/t/about-pets-prize-challenge-next-year/9874 "2024-01-03T15:07:46Z")
hi, i want to know, will this competition be held again in 2024? --- ## [PETs Challenge - Data Availability](https://community.drivendata.org/t/pets-challenge-data-availability/7877)
**Author:** [@sony.rajan](https://community.drivendata.org/u/sony.rajan)\ **Replies:** 21\ **Last updated:** [November 22, 2023, 10:02pm UTC](https://community.drivendata.org/t/pets-challenge-data-availability/7877 "2023-11-22T22:02:09Z")
There is no public data for this yet but the Track page says that you will be provided datasets representing synthetic population data that includes: a social contact network, capturing when and where any two people co… --- ## [Labels test data of evaluation sets](https://community.drivendata.org/t/labels-test-data-of-evaluation-sets/8628)
**Author:** [@JasperDekoninck](https://community.drivendata.org/u/JasperDekoninck)\ **Replies:** 3\ **Last updated:** [February 14, 2023, 3:13am UTC](https://community.drivendata.org/t/labels-test-data-of-evaluation-sets/8628 "2023-02-14T03:13:09Z")
Hi! Would it be possible for the red teams to also get access to the correct labels for the test datasets? I noticed in the Financial Crime dataset that this is missing. I assume this will also be the case for the Pande… --- ## [Privacy Scenarios Blue Teams](https://community.drivendata.org/t/privacy-scenarios-blue-teams/8588)
**Author:** [@JasperDekoninck](https://community.drivendata.org/u/JasperDekoninck)\ **Replies:** 1\ **Last updated:** [February 7, 2023, 6:45pm UTC](https://community.drivendata.org/t/privacy-scenarios-blue-teams/8588 "2023-02-07T18:45:01Z")
Hi! Blue teams have to provide three privacy scenarios. Do we need to attack all three scenarios? Or does the score eg. get lower if we implement an attack that is only applicable/good for the lowest privacy setting? We… --- ## [Datasets Track A](https://community.drivendata.org/t/datasets-track-a/8567)
**Author:** [@JasperDekoninck](https://community.drivendata.org/u/JasperDekoninck)\ **Replies:** 1\ **Last updated:** [February 1, 2023, 7:02pm UTC](https://community.drivendata.org/t/datasets-track-a/8567 "2023-02-01T19:02:53Z")
Hi! We just noticed in the data download page (for Red Teams) that there is a new dataset for Track A (not sure since when it has been there). The old dataset is still there and we have noticed that their characteristic… --- ## [Submission requirements Red Phase](https://community.drivendata.org/t/submission-requirements-red-phase/8543)
**Author:** [@JasperDekoninck](https://community.drivendata.org/u/JasperDekoninck)\ **Replies:** 3\ **Last updated:** [January 31, 2023, 10:12pm UTC](https://community.drivendata.org/t/submission-requirements-red-phase/8543 "2023-01-31T22:12:55Z")
Hi! I had a couple of questions about the submission format for the red phase participants. The outline given now, is an outline that assumes only one attack is implemented. However, since we have to evaluate multiple s… --- ## [Specifying different hyperparameters for different federated scenarios](https://community.drivendata.org/t/specifying-different-hyperparameters-for-different-federated-scenarios/8513)
**Author:** [@kzliu](https://community.drivendata.org/u/kzliu)\ **Replies:** 1\ **Last updated:** [January 25, 2023, 9:59pm UTC](https://community.drivendata.org/t/specifying-different-hyperparameters-for-different-federated-scenarios/8513 "2023-01-25T21:59:06Z")
Hi @jayqi, Would it be possible (or encouraged) to specify different hyperparameters (e.g. number of training rounds) for different federated scenarios scenarioXX? The main motivation is that the best performing setting… --- ## [Track B: caching preprocessed data](https://community.drivendata.org/t/track-b-caching-preprocessed-data/8512)
**Author:** [@kzliu](https://community.drivendata.org/u/kzliu)\ **Replies:** 4\ **Last updated:** [January 25, 2023, 5:54pm UTC](https://community.drivendata.org/t/track-b-caching-preprocessed-data/8512 "2023-01-25T17:54:42Z")
Hi @jayqi, Are solutions allowed to cache preprocessed data (features) to disk? If so, are there any limit on the disk space? Specifically, from Track B: quick clarification regarding train/test folders in the runtime … --- ## [Announcement: Update to Submission Deadline](https://community.drivendata.org/t/announcement-update-to-submission-deadline/8511)
**Author:** [@jayqi](https://community.drivendata.org/u/jayqi)\ **Replies:** 0\ **Last updated:** [January 24, 2023, 8:22pm UTC](https://community.drivendata.org/t/announcement-update-to-submission-deadline/8511 "2023-01-24T20:22:42Z")
Hello PETs Prize Challenge participants, Thanks to all teams for participating in Phase 2 of the challenge. Challenge organizers have decided to keep Phase 2 submissions open for an additional two days, through January … --- ## [Bank name in transaction is not found](https://community.drivendata.org/t/bank-name-in-transaction-is-not-found/8503)
**Author:** [@panheng](https://community.drivendata.org/u/panheng)\ **Replies:** 4\ **Last updated:** [January 24, 2023, 6:55pm UTC](https://community.drivendata.org/t/bank-name-in-transaction-is-not-found/8503 "2023-01-24T18:55:58Z")
I got an exception from the official runtime: Generally, in our fl solution, each client will send to swift client the list of banks it has. And then the swift client can build a dict mapping bank name to client id. … --- ## [Selecting submissions after the deadline](https://community.drivendata.org/t/selecting-submissions-after-the-deadline/8506)
**Author:** [@kzliu](https://community.drivendata.org/u/kzliu)\ **Replies:** 1\ **Last updated:** [January 24, 2023, 3:10pm UTC](https://community.drivendata.org/t/selecting-submissions-after-the-deadline/8506 "2023-01-24T15:10:20Z")
Hi @jayqi, Would it be possible to select submissions that started before the deadline but finished after the deadline? For example, if I have only one completed submission now, but wanted to add another one with differ… --- ## [Track B: execution failed for federated without messages](https://community.drivendata.org/t/track-b-execution-failed-for-federated-without-messages/8501)
**Author:** [@kzliu](https://community.drivendata.org/u/kzliu)\ **Replies:** 1\ **Last updated:** [January 24, 2023, 6:44am UTC](https://community.drivendata.org/t/track-b-execution-failed-for-federated-without-messages/8501 "2023-01-24T06:44:27Z")
Hi @jayqi! I’m observing a weird failure for my federated smoke test (at timestep 2023-01-23 20:23:47 UTC). My code has successfully ran till the testing stage of the 3rd scenario, and then it abruptly failed at withou… --- ## [Version consistency between CUDA, PyTorch and other related modules](https://community.drivendata.org/t/version-consistency-between-cuda-pytorch-and-other-related-modules/8462)
**Author:** [@ryokawa](https://community.drivendata.org/u/ryokawa)\ **Replies:** 9\ **Last updated:** [January 24, 2023, 3:30am UTC](https://community.drivendata.org/t/version-consistency-between-cuda-pytorch-and-other-related-modules/8462 "2023-01-24T03:30:46Z")
Hi, When I was checking the versions of CUDA and PyTorch in the docker container that is used in pets-prize-challenge-runtime, I noticed that the version numbers returned through various interfaces seem to be inconsiste… --- ## [Understanding Final Submission Due](https://community.drivendata.org/t/understanding-final-submission-due/8499)
**Author:** [@yizhewan](https://community.drivendata.org/u/yizhewan)\ **Replies:** 1\ **Last updated:** [January 24, 2023, 3:07am UTC](https://community.drivendata.org/t/understanding-final-submission-due/8499 "2023-01-24T03:07:11Z")
Hi, So the final submission due is Jan 24 23:59:59 GMT. My question is, do we need to finish running the code on online judge before that due as well or just get it running? --- ## [Track B output format](https://community.drivendata.org/t/track-b-output-format/8498)
**Author:** [@hhcho](https://community.drivendata.org/u/hhcho)\ **Replies:** 1\ **Last updated:** [January 24, 2023, 1:44am UTC](https://community.drivendata.org/t/track-b-output-format/8498 "2023-01-24T01:44:05Z")
Hi @jayqi, We have some last minute clarification questions regarding the output format for Track B (pandemic): Does the evaluation pipeline depend on the fact that the predicted scores are between 0 and 1? Or will we… --- ## [Resource issue when submitting for Track B federated](https://community.drivendata.org/t/resource-issue-when-submitting-for-track-b-federated/8496)
**Author:** [@kzliu](https://community.drivendata.org/u/kzliu)\ **Replies:** 5\ **Last updated:** [January 23, 2023, 6:39pm UTC](https://community.drivendata.org/t/resource-issue-when-submitting-for-track-b-federated/8496 "2023-01-23T18:39:12Z")
Hi, I’m running into the following errors with my federated submission for Track B Pandemic: ... Error: No available node types can fulfill resource request {'CPU': 5.0, 'GPU': 1.0}. Add suitable node types to this clus… --- ## [ReUse of train-data generated features for test phase](https://community.drivendata.org/t/reuse-of-train-data-generated-features-for-test-phase/8495)
**Author:** [@kingaj12](https://community.drivendata.org/u/kingaj12)\ **Replies:** 4\ **Last updated:** [January 23, 2023, 6:21pm UTC](https://community.drivendata.org/t/reuse-of-train-data-generated-features-for-test-phase/8495 "2023-01-23T18:21:23Z")
In the phase-one notebook, the features generated on training data (\`sender\_currency\_freq’ etc) were used both to fit the RF or XGBoost models and for the prediction on the test data. For phase-two: is this reuse of tra… --- ## [Cannot run local eval for fl solutions, fincrime](https://community.drivendata.org/t/cannot-run-local-eval-for-fl-solutions-fincrime/8492)
**Author:** [@panheng](https://community.drivendata.org/u/panheng)\ **Replies:** 1\ **Last updated:** [January 23, 2023, 3:09am UTC](https://community.drivendata.org/t/cannot-run-local-eval-for-fl-solutions-fincrime/8492 "2023-01-23T03:09:28Z")
I was using the following command. SUBMISSION\_TRACK=fincrime SUBMISSION\_TYPE=federated GITHUB\_ACTIONS\_NO\_TTY=true make test-submission The tested submission was the example source code. But after configure\_fit complete… --- ## [Clarification on federated pandemic model state storage](https://community.drivendata.org/t/clarification-on-federated-pandemic-model-state-storage/8489)
**Author:** [@denis.broadinstitute](https://community.drivendata.org/u/denis.broadinstitute)\ **Replies:** 3\ **Last updated:** [January 23, 2023, 1:34am UTC](https://community.drivendata.org/t/clarification-on-federated-pandemic-model-state-storage/8489 "2023-01-23T01:34:50Z")
Hi @jayqi, We’ve followed this guideline so far: Your code must use the intended Flower APIs for client–server communication. Your code must only use the provided client and server directories for saving and loading … --- ## [Cannot Select Final Submission](https://community.drivendata.org/t/cannot-select-final-submission/8484)
**Author:** [@yizhewan](https://community.drivendata.org/u/yizhewan)\ **Replies:** 4\ **Last updated:** [January 22, 2023, 9:43pm UTC](https://community.drivendata.org/t/cannot-select-final-submission/8484 "2023-01-22T21:43:58Z")
Hi, I made two submissions for the centralized setting. However, I’m not able to select any of them as my final submission… When I click on any of the checkbox, I receive error saying I’ve selected a submission already.… --- ## [Final Execution Time Limit - FinCrime Federated](https://community.drivendata.org/t/final-execution-time-limit-fincrime-federated/8468)
**Author:** [@jimking100](https://community.drivendata.org/u/jimking100)\ **Replies:** 4\ **Last updated:** [January 22, 2023, 2:08am UTC](https://community.drivendata.org/t/final-execution-time-limit-fincrime-federated/8468 "2023-01-22T02:08:04Z")
Hi, It appears that the 9 hour time limit for the federated final submission is not enough time to run my solution. It runs fine in the smoke test and gets almost to the end of scenario 2 before timing out. I believe … --- ## [Final Submission - Pandemic Centralized](https://community.drivendata.org/t/final-submission-pandemic-centralized/8481)
**Author:** [@jimking100](https://community.drivendata.org/u/jimking100)\ **Replies:** 1\ **Last updated:** [January 20, 2023, 11:42pm UTC](https://community.drivendata.org/t/final-submission-pandemic-centralized/8481 "2023-01-20T23:42:02Z")
Hi, I receive the following error in the log when I submit code for Pandemic Centralized final solution: /opt/conda/envs/condaenv/.tmpuswpugnr: line 3: 40 Killed python main\_centralized\_train.py ER… --- ## [Documentation Submission Query](https://community.drivendata.org/t/documentation-submission-query/8478)
**Author:** [@sarahrhalem](https://community.drivendata.org/u/sarahrhalem)\ **Replies:** 3\ **Last updated:** [January 20, 2023, 9:41pm UTC](https://community.drivendata.org/t/documentation-submission-query/8478 "2023-01-20T21:41:51Z")
Hello, We have a question on documentation submission for Phase II. We are allowed 4 additional pages for updates excluding references. Are we also allowed to include a Title page, Contents page and a page listing team … --- ## [Uniqueness of MessageId/UETR](https://community.drivendata.org/t/uniqueness-of-messageid-uetr/8474)
**Author:** [@yizhewan](https://community.drivendata.org/u/yizhewan)\ **Replies:** 3\ **Last updated:** [January 20, 2023, 9:42am UTC](https://community.drivendata.org/t/uniqueness-of-messageid-uetr/8474 "2023-01-20T09:42:42Z")
Hi, May I know if the MessageId is a unique identifier of the transactions in the final evaluation sets? Is UETR unique too? Additionally, will the train and test transactions have overlapping UETR and/or MessageId? Tha… --- ## [Long wait for smoke test submission](https://community.drivendata.org/t/long-wait-for-smoke-test-submission/8464)
**Author:** [@yizhewan](https://community.drivendata.org/u/yizhewan)\ **Replies:** 1\ **Last updated:** [January 19, 2023, 4:48pm UTC](https://community.drivendata.org/t/long-wait-for-smoke-test-submission/8464 "2023-01-19T16:48:35Z")
Dear organizers, Today I noticed the wait time for smoke test is significantly longer: my code submission stuck at “Starting: The execution environment for your code is starting up” for more than 40 mins. Is this wait t… --- ## [Extension for Paper Writing](https://community.drivendata.org/t/extension-for-paper-writing/8434)
**Author:** [@yizhewan](https://community.drivendata.org/u/yizhewan)\ **Replies:** 2\ **Last updated:** [January 18, 2023, 6:25pm UTC](https://community.drivendata.org/t/extension-for-paper-writing/8434 "2023-01-18T18:25:53Z")
Hi organizers, I’m writing to ask if it is possible to move the paper submission due slightly later than the online code submission due. On one hand, We want to try various techniques to push the performance as much as … --- ## [Announcement: Making Final Submissions](https://community.drivendata.org/t/announcement-making-final-submissions/8461)
**Author:** [@jayqi](https://community.drivendata.org/u/jayqi)\ **Replies:** 0\ **Last updated:** [January 18, 2023, 6:12pm UTC](https://community.drivendata.org/t/announcement-making-final-submissions/8461 "2023-01-18T18:12:48Z")
Hello PETs Prize Challenge participants, Welcome to the home stretch! Final submissions eligible for Phase 2 prizes are due by January 24, 2023 January 26, 2023 at 23:59:59 UTC . (Please note the time zone!) See below f… --- ## [Sending local model weights to clients](https://community.drivendata.org/t/sending-local-model-weights-to-clients/8433)
**Author:** [@mmemon](https://community.drivendata.org/u/mmemon)\ **Replies:** 7\ **Last updated:** [January 17, 2023, 5:26pm UTC](https://community.drivendata.org/t/sending-local-model-weights-to-clients/8433 "2023-01-17T17:26:20Z")
Hello, As per our implementation, we require to send model weights from an individual local client model to a server model in order to identify the models to select for training in the subsequent rounds(trust and cluste… --- ## [Track B: clarification over target label](https://community.drivendata.org/t/track-b-clarification-over-target-label/8438)
**Author:** [@domdathan](https://community.drivendata.org/u/domdathan)\ **Replies:** 2\ **Last updated:** [January 13, 2023, 4:06pm UTC](https://community.drivendata.org/t/track-b-clarification-over-target-label/8438 "2023-01-13T16:06:45Z")
Hi, Apologies if this has been asked previously, but could you clarify what the target label is for the pandemic track. On the data overview page it is described in two places as: Your task is to predict a risk score… [Next page](https://community.drivendata.org/c/pets-prize-federated-learning/88.md?page=1)