Skip to main content

timecopilot-kumo-forecast

Inference-only packaging of NVIDIA Kumo-Forecast (nvidia/Kumo-Forecast on Hugging Face), used by FoundationForecast.

Upstream is clone-only and installs top-level modules named sdk, backbone and model. This package ships the same backbone under the single kumo_forecast namespace, without the anomaly, interpretability, DARR and fine-tuning code, and without downloading the MOMENT-1-large weights (the Kumo checkpoint contains every weight).

import numpy as np

from kumo_forecast import load_kumo_forecast

model = load_kumo_forecast()  # downloads nvidia/Kumo-Forecast
y = np.sin(np.arange(300) * 2 * np.pi / 24)[None]  # [n_series, length], NaN allowed
forecast = model.forecast(y, h=24)  # [n_series, 24]

The checkpoint has a fixed 512-step context and a native horizon of 72; shorter series are left-padded and masked, longer ones truncated, and horizons above 72 use an autoregressive rollout. Forecasts are point forecasts only.

Licensed under Apache-2.0. See NOTICE.

Metadata

Release files for timecopilot-kumo-forecast 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for timecopilot-kumo-forecast 0.1.0
File Size Uploaded Provenance Actions
timecopilot_kumo_forecast-0.1.0.tar.gz 137.6 kB Without provenance Details for timecopilot_kumo_forecast-0.1.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for timecopilot-kumo-forecast 0.1.0
File Interpreter ABI Platform Actions
timecopilot_kumo_forecast-0.1.0-py3-none-any.whl Python 3 none any Details for timecopilot_kumo_forecast-0.1.0-py3-none-any.whl

Total release size: 155.1 kB

Release files / timecopilot_kumo_forecast-0.1.0.tar.gz

Download URL timecopilot_kumo_forecast-0.1.0.tar.gz
Size 137.6 kB
Tags Source
SHA-256 checksum
How to use checksums
c66ebe7d9be0fe9ee4f8dcfbb0b8a2ba0df6be844b6b3c7add7ac128575efb7c
BLAKE2b-256 checksum
How to use checksums
2a228cd247d191e4061a051acfc8f5c8524e283f493ffd1db4b44bb0f7f552e9
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.7.8

Release files / timecopilot_kumo_forecast-0.1.0-py3-none-any.whl

Download URL timecopilot_kumo_forecast-0.1.0-py3-none-any.whl
Size 17.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
6f8f7482723481773a52be215b6daf597edbdafd6fff253495f526267089684e
BLAKE2b-256 checksum
How to use checksums
d1ed4345fde5b055df13e0b5a0a2beed831dbd5f05fdebab4abfdd3a5c2ce88b
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.7.8

Security for timecopilot-kumo-forecast 0.1.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page