Skip to content
GluonTS - Probabilistic Time Series Modeling in Python
Branch: master
Clone or download
SSappy and lostella Adding option to rescale time series instead of clipping them (#114)
* Adding option to rescale time series instead of clipping them

* Adding tests for complex seasonal and clipping integer time series to the right range

* Update test/dataset/artificial/test_complex_seasonal.py

Co-Authored-By: Lorenzo Stella <lorenzostella@gmail.com>

* Reducing the number of tests and commenting the values mapping

* Removing randomness in tests
Latest commit 64eeba1 Jun 18, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.devtools/githooks remove flake8 from githook (#109) Jun 14, 2019
.github Initial code commit May 17, 2019
ci Disable warnings-as-failure for doc builds on master (#72) Jun 17, 2019
docs update 404 image (#90) Jun 7, 2019
env Website and CI (#10) May 30, 2019
evaluations small table formatting and script fix (#84) Jun 7, 2019
examples add datasets (#70) Jun 5, 2019
src/gluonts Adding option to rescale time series instead of clipping them (#114) Jun 18, 2019
test Adding option to rescale time series instead of clipping them (#114) Jun 18, 2019
.gitignore Migrate project and setup.py to `src/`-based layout (#60) Jun 4, 2019
.pylintrc Website and CI (#10) May 30, 2019
.travis.yml Updated travis password. (#102) Jun 13, 2019
CODE_OF_CONDUCT.md Initial code commit May 17, 2019
CONTRIBUTING.md remove flake8 from githook (#109) Jun 14, 2019
LICENSE Initial code commit May 17, 2019
Makefile Add --no-dir-entries to zip file creation to prune empty directories (#… Jun 5, 2019
NOTICE Add Apache 2.0 license file headers and update NOTICE (#48) Jun 3, 2019
README.md add bibtex entry to readme (#119) Jun 18, 2019
codecov.yml Website and CI (#10) May 30, 2019
conftest.py Website and CI (#10) May 30, 2019
contributor.rst Update contributors.rst (#47) Jun 3, 2019
dev_setup.sh remove flake8 from githook (#109) Jun 14, 2019
pyproject.toml Updated black to version 19. (#118) Jun 17, 2019
pytest.ini Website and CI (#10) May 30, 2019
requirements-extras-prophet.txt Add extra requirements for R forecast and Prophet (#27) May 29, 2019
requirements-extras-r.txt Fix rpy2 version requirement (#89) Jun 7, 2019
requirements-setup.txt Updated black to version 19. (#118) Jun 17, 2019
requirements-test.txt Update requirements files (#55) Jun 4, 2019
requirements.txt Updating pydantic to `0.28`. (#108) Jun 14, 2019
setup.cfg Update setup.cfg (#4) Jun 13, 2019
setup.py Migrate project and setup.py to `src/`-based layout (#60) Jun 4, 2019

README.md

GluonTS - Probabilistic Time Series Modeling in Python

GluonTS is a Python toolkit for probabilistic time series modeling, built around Apache MXNet (incubating).

GluonTS provides utilities for loading and iterating over time series datasets, state of the art models ready to be trained, and building blocks to define your own models and quickly experiment with different solutions.

Installation

GluonTS requires Python 3.6, and the easiest way to install it is via pip:

pip install gluonts

Quick start guide

This simple example illustrates how to train a model from GluonTS on some data, and then use it to make predictions. As a first step, we need to collect some data: in this example we will use the volume of tweets mentioning the AMZN ticker symbol.

import pandas as pd
url = "https://raw.githubusercontent.com/numenta/NAB/master/data/realTweets/Twitter_volume_AMZN.csv"
df = pd.read_csv(url, header=0, index_col=0)

The first 100 data points look like follows:

import matplotlib.pyplot as plt
df[:100].plot(linewidth=2)
plt.grid(which='both')
plt.show()

Data

We can now prepare a training dataset for our model to train on. Datasets in GluonTS are essentially iterable collections of dictionaries: each dictionary represents a time series with possibly associated features. For this example, we only have one entry, specified by the "start" field which is the timestamp of the first datapoint, and the "target" field containing time series data. For training, we will use data up to midnight on April 5th, 2015.

from gluonts.dataset.common import ListDataset
training_data = ListDataset(
    [{"start": df.index[0], "target": df.value[:"2015-04-05 00:00:00"]}],
    freq = "5min"
)

A forecasting model in GluonTS is a predictor object. One way of obtaining predictors is by training a correspondent estimator. Instantiating an estimator requires specifying the frequency of the time series that it will handle, as well as the number of time steps to predict. In our example we're using 5 minutes data, so freq="5min", and we will train a model to predict the next hour, so prediction_length=12. We also specify some minimal training options.

from gluonts.model.deepar import DeepAREstimator
from gluonts.trainer import Trainer

estimator = DeepAREstimator(freq="5min", prediction_length=12, trainer=Trainer(epochs=10))
predictor = estimator.train(training_data=training_data)

During training, useful information about the progress will be displayed. To get a full overview of the available options, please refer to the documentation of DeepAREstimator (or other estimators) and Trainer.

We're now ready to make predictions: we will forecast the hour following the midnight on April 15th, 2015.

test_data = ListDataset(
    [{"start": df.index[0], "target": df.value[:"2015-04-15 00:00:00"]}],
    freq = "5min"
)

from gluonts.dataset.util import to_pandas

for test_entry, forecast in zip(test_data, predictor.predict(test_data)):
    to_pandas(test_entry)[-60:].plot(linewidth=2)
    forecast.plot(color='g', prediction_intervals=[50.0, 90.0])
plt.grid(which='both')

Forecast

Note that the forecast is displayed in terms of a probability distribution: the shaded areas represent the 50% and 90% prediction intervals, respectively, centered around the median (dark green line).

Further examples

The following are good entry-points to understand how to use many features of GluonTS:

The following modules illustrate how custom models can be implemented:

Contributing

If you wish to contribute to the project, please refer to our contribution guidelines.

Citing

If you use GluonTS in a scientific publication, we encourage you to add the following reference to the associated paper:

@article{gluonts,
  title={{GluonTS: Probabilistic Time Series Modeling in Python}},
  author={Alexandrov, A. and Benidis, K. and Bohlke-Schneider, M. and
          Flunkert, V. and Gasthaus, J. and Januschowski, T. and Maddix, D. C.
          and Rangapuram, S. and Salinas, D. and Schulz, J. and Stella, L. and
          Türkmen, A. C. and Wang, Y.},
  journal={arXiv preprint arXiv:1906.05264},
  year={2019}
}
You can’t perform that action at this time.