A probabalistic programming ML tool for science
Project description
pylace
Python bindings to lace
Install
Install lates from PyPI
$ python3 -m pip install pylace
Install latest from GitHub
Building from source requires the Rust compiler (git it here).
$ python3 -m pip install git+https://git@github.com/promised-ai/lace.git#egg=pylace&subdirectory=pylace
Use
Note that the engine currently only supports loading from an existing metadata file. The following lace functions are supported:
- rowsim
- depprob
- predict
- logp
- simulate
- append_rows
- update
import lace
# The required files can be found here: https://github.com/promised-ai/lace/tree/master/pylace/lace/resources/datasets/satellites
engine = lace.Engine(data_source="data.csv", codebook="codebook.yaml")
# Train the model for 10_000 steps
engine.update(10_000)
# Predict the orbit's class based on the orbit's period.
engine.predict('Class_of_Orbit', given={'Period_minutes': 1436.0})
# ('GEO', 0.13583714831550336)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
pylace-0.1.0a1.tar.gz
(885.0 kB
view hashes)
Built Distribution
Close
Hashes for pylace-0.1.0a1-cp310-cp310-macosx_10_7_x86_64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 4bf9b5676bfde020288cdfd7c934e60e5facd31049d0ff53781e70a072e2e1a8 |
|
MD5 | 2fe979eaac604a8de2457002eb361bd0 |
|
BLAKE2b-256 | 409c478045f25691502368f08362698c556b09060cddaf0180e10d95ff740b35 |