Skip to main content

Deep learning library for time series forecasting based on PyTorch.

Project description

DeepTimeSeries

Last update Oct.19, 2022

Deep learning library for time series forecasting based on PyTorch. It is under development and the first released version will be announced soon.

Why DeepTimeSeries?

DeepTimeSeries is inspired by libraries such as Darts and PyTorch Forecasting. So why was DeepTimeSeries developed?

The design philosophy of DeepTimeSeries is as follows:

We present logical guidelines for designing various deep learning models for time series forecasting

Our main target users are intermediate-level users who need to develop deep learning models for time series prediction. We provide solutions to many problems that deep learning modelers face because of the uniqueness of time series data.

We additionally implement a high-level API, which allows comparatively beginners to use models that have already been implemented.

Supported Models

Model Target features Non-target features Deterministic Probabilistic
MLP o o o o
Dilated CNN o o o o
Vanilla RNN o o o o
LSTM o o o o
GRU o o o o
Transformer o o o o

Documentation

https://bet-lab.github.io/DeepTimeSeries/

Project details


Download files

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

Source Distribution

deep_time_series-0.1.2.tar.gz (11.1 kB view hashes)

Uploaded Source

Built Distribution

deep_time_series-0.1.2-py3-none-any.whl (16.6 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page