Additional Info
| Field |
Value |
| Source |
https://github.com/The-EPISERVE-Consortium/model__prediction__generic__timesfm
|
| Last Updated |
September 2, 2026, 21:08 (UTC)
|
| Created |
September 2, 2026, 20:27 (UTC)
|
| algorithm |
|
| dataset_type |
model |
| docker_image |
ghcr.io/the-episerve-consortium/model__prediction__generic__timesfm |
| docker_image_created |
2026-09-02T19:00:09.510881982Z |
| docker_tag |
latest |
| input_format |
|
| lead_researcher |
|
| model_parameters |
[{"@type": "PropertyValue", "name": "history_length", "description": "Number of rows to use as model context. Window ends at total_rows - prediction_offset. Minimum 3 (the model's minimum context). Values above 2048 use only the most recent 2048 rows.", "valueRequired": true, "minValue": 3}, {"@type": "PropertyValue", "name": "prediction_length", "description": "Number of steps to forecast ahead.", "valueRequired": true, "minValue": 1, "maxValue": 512}, {"@type": "PropertyValue", "name": "prediction_offset", "description": "Rows to skip at the end before the history window. Use for back-testing against known data.", "valueRequired": false, "minValue": 0, "value": 0}] |
| model_qid |
Q7611536028271 |
| output_format |
|
| paper_doi |
|
You can access all data via a web API using e.g. Python or curl.
Python
import requests
dataset_id = "q7611536028271"
url = "https://data.episerve.zib.de/api/3/action/package_show"
response = requests.get(url, params={"id": dataset_id})
dataset = response.json()["result"]
print(dataset["title"])
for resource in dataset["resources"]:
print(resource["name"], resource["url"])
curl
curl "https://data.episerve.zib.de/api/3/action/package_show?id=q7611536028271"