Space-time prediction with sparse and irregular space-time multi-timeserie.

Project description

steams

Space-time prediction with sparse and irregular space-time multi-timeseries.

Models presented in this packages are using an adaptive distance attention mechanism. The weight of the attention are based either on the Ordinary Kriging equation system or the Nadaraya-Watson Kernel.

Install from PyPi

pip install steams

install from source

cd /tmp
git clone https://git.nilu.no/aqdl/steams_pkg.git
cd steams_pkg
pip3 install -e .

Package 'steams' has been tested on python 3.8 and 3.9

Running 'steams' with CUDA (v11.3), requires a manual installation of pytorch:

pip3 install torch==1.11.0+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html

Project details

Release history Release notifications | RSS feed

This version

0.18

Feb 13, 2023

0.17

Jan 12, 2023

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

steams-0.18.tar.gz (15.1 kB view details)

Uploaded Feb 13, 2023 Source

Built Distribution

steams-0.18-py3-none-any.whl (16.8 kB view details)

Uploaded Feb 13, 2023 Python 3

File details

Details for the file steams-0.18.tar.gz.

File metadata

Download URL: steams-0.18.tar.gz
Upload date: Feb 13, 2023
Size: 15.1 kB
Tags: Source
Uploaded using Trusted Publishing? No
Uploaded via: twine/4.0.2 CPython/3.8.10

File hashes

Hashes for steams-0.18.tar.gz
Algorithm	Hash digest
SHA256	`cf32045a1dad3d86abbfa3f923b40e18811824112b9f726577f47afd716c0605`
MD5	`bdb0f8948990f8a784e8c2ab1eafcbd1`
BLAKE2b-256	`1b11e0f5f87a46b3416be39f7f0543f73d6d68a02afb3ce8ce3142df63ec78de`

See more details on using hashes here.

File details

Details for the file steams-0.18-py3-none-any.whl.

File metadata

Download URL: steams-0.18-py3-none-any.whl
Upload date: Feb 13, 2023
Size: 16.8 kB
Tags: Python 3
Uploaded using Trusted Publishing? No
Uploaded via: twine/4.0.2 CPython/3.8.10

File hashes

Hashes for steams-0.18-py3-none-any.whl
Algorithm	Hash digest
SHA256	`7763699f352274d35acda1ea9985303df7cc1a35e028d38a428e9edbf33cc76c`
MD5	`821ace87057fccc2723d24fc178d082d`
BLAKE2b-256	`6549ea8ff09a6932d43e174b2c5317747295ff2cd904a511572f3348e623b05b`