Mind Foundry Horizon Client
Project description
Mind Foundry Horizon Client
This package contains a python client that can be used to interact with an instance of Mind Foundry Horizon. The client exposes a familiar interface that allows you to tackle forecasting problems with ease.
from mindfoundry.client.horizon import Connection
from mindfoundry.client.horizon.models import HorizonForecaster
# Generate an API key in the Horizon dashboard
connection = Connection(
base_url="your Horizon instance",
api_key="your api key",
)
# Create a new forecasting model. See the complete documentation
# for a more in-depth explanation of each argument.
model = HorizonForecaster(
connection=connection,
name="My Forecaster",
targets=["target1", "target2", ...],
horizons=[1, 2, 3],
refinement=False,
)
# Train a model on some training data (a pandas dataframe)
model.fit(training_data)
# Make a prediction on new data
prediction = model.predict(new_data)
print(prediction.as_df().head())
# Save the model for later use
model.save("/path/to/model/file")
# Re-load the model
model = HorizonForecaster.load("/path/to/model/file", connection=connection)
# The model can be updated with new data in the same format
model.update(new_data_2)
new_prediction = model.predict()
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
Built Distribution
Close
Hashes for mindfoundry.client.horizon-2.12.3.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | d48deb293cbcba7c6d03487cd3fb68768e23c38a96113ed44a27d1cf055fe857 |
|
MD5 | 119d08e13dd71489a629e342d156d3bf |
|
BLAKE2b-256 | 80f670f1936061c7438b86f906ac217f49d29c98abd8ed0ab913bdc7f0861188 |
Close
Hashes for mindfoundry.client.horizon-2.12.3-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 006ac6ce501a5772bed7d72bcc65653dc2749dad9946ff9b376e6c7a51740f80 |
|
MD5 | 3bbd131cb2ee2a1d7296f86edd20de53 |
|
BLAKE2b-256 | c7572ce6be84b6318880f2d5cb282d702a3e65a8fd647f25406b72ccc5ded6e0 |