Austin 911 response time
Predict late first-unit arrival using information available when the call-taker's dispatch screen opens.
About this release
This release turns City of Austin call records into a prediction task with temporal training, evaluation and holdout splits. The construction code, qualification checks and a reproducible baseline travel with the dataset.
The target uses the source's response_time field. The threshold was selected using the training period and then frozen. Prediction time and the outcome's clock are distinct: the source measures response time from the call being answered, while response_datetime records the opening of the call-taker's dispatch screen.
The recorded baseline demonstrates the runner's training and prediction contract on this artifact. It is not an agent or harness comparison.
Splits
| Split | Rows | Positive labels |
|---|---|---|
| eval | 285,665 | 109,860 (38.46%) |
| holdout | 191,791 | 72,450 (37.78%) |
| train | 572,180 | 231,025 (40.38%) |
Qualification and baseline
PASSED - 33 checks, 0 failed, on the full 1,049,636-row artifact, on the worker Gate 1.3.0.
one baseline only. No agent or harness comparison was run, and none is implied - that is outside this publication milestone.
Use this version
from datasets import load_dataset
ds = load_dataset('earino/austin-911-response', revision='v2026.09')The version links and checksums identify this artifact. Exclude the labelled holdout from the agent's workspace when running an evaluation.
Files and checksums
| Download | Bytes | SHA-256 |
|---|---|---|
| eval.csv | 35,801,537 | 241dee5eb84cfdd5e1101a3453d2af4dea9bdf9dfca549d83323584a6e442d7a |
| holdout.csv | 24,075,321 | 738ff7ec245302ff2a8dcfb1be95dd622b6d8e7acaf8e9a025b1823a430cab7a |
| meta.json | 2,121 | 5dc8aed06c157a35955e7913748609cf6c42be2b2d144f405de41240dd9d45fb |
| quality.json | 3,434 | a7f9c99c5d0c9465e28ef66eee9ffa9c7f15c53d094148567c2ff48c961241a7 |
| train.csv | 71,546,966 | 6fc37e034447b2e982d737f6d62d0da9a24584d89d25799d5d882ba584604434 |
Artifact e4598317e406984fa590aacc5e7aff675867578c51ae1a84ebf6279bc42c3328
Limitations
- This is a derived research dataset, not an official City of Austin product or a recommendation about emergency response.
- The source outcome clock and prediction timestamp differ. Read the documented clock-offset measurements before interpreting the target.
- The holdout is downloadable. Keep its labels outside the evaluated agent's workspace during an evaluation.
Source and terms
Austin Open Data - APD 911 Calls for Service 2023-2026
City of Austin, APD 911 Calls for Service 2023-2026, Austin Open Data, https://datahub.austintexas.gov/d/e687-fx2y, accessed 2026-09-18
Source: LicenseRef-Public-Domain. Compilation: CC0-1.0. Code and documentation: MIT.
Our contribution: row selection, the derived binary label, the temporal partition and the packaging. Does not cover or relicense the City of Austin's data.
Suggested citation
City of Austin. APD 911 Calls for Service 2023-2026. Austin Open Data. Public Domain. Accessed 2026-09-18. Derived dataset: Austin 911 response time prediction dataset, version 2026.09, artifact e4598317e406984fa590aacc5e7aff675867578c51ae1a84ebf6279bc42c3328.