https://github.com/pratham16cse/DualTPP

Code for "Long Horizon Forecasting With Temporal Point Processes", WSDM 2021
https://github.com/pratham16cse/DualTPP

event-sequence forecasting long-term-forecasting multi-view-learning point-process temporal-modeling temporal-point-processes

Last synced: about 1 month ago
JSON representation

Code for "Long Horizon Forecasting With Temporal Point Processes", WSDM 2021

• Host: GitHub
• URL: https://github.com/pratham16cse/DualTPP
• Owner: pratham16cse
• Created: 2020-09-20T07:45:12.000Z (almost 4 years ago)
• Default Branch: master
• Last Pushed: 2022-02-05T10:46:26.000Z (over 2 years ago)
• Last Synced: 2024-05-21T11:49:50.821Z (4 months ago)
• Topics: event-sequence, forecasting, long-term-forecasting, multi-view-learning, point-process, temporal-modeling, temporal-point-processes
• Language: Python
• Homepage: https://arxiv.org/abs/2101.02815
• Size: 238 MB
• Stars: 21
• Watchers: 2
• Forks: 5
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-time-series - Code

README

        
# Long Horizon Forecasting With Temporal Point Processes

![DualTPP Diagram](DualTPP_diagram.png)

This is the code produced as part of the paper _Long Horizon Forecasting With Temporal Point Processes_ 

> "Long Horizon Forecasting With Temporal Point Processes"

> Prathamesh Deshpande, Kamlesh Marathe, Abir De, Sunita Sarawagi. WSDM 2021. [arXiv:2101.02815](https://arxiv.org/abs/2101.02815)

## Packages needed

Specified in [requirements](requirements.txt).

## Dataset Download

We have provided all the datasets used in our experiments [here](https://drive.google.com/drive/folders/1b1KUwkeIqIViPZoRZzbPAzKeNn7P1OD-?usp=sharing).

Please download the `data/` folder add place it in the [DualTPP](https://github.com/pratham16cse/DualTPP) directory.

## Experiment execution

To run the code to reproduce the results, please use this [script](script.sh) \[ Under development, more datasets will be soon added to the script\].

## Output

All the outputs will be stored in the `` directory.

The numbers reported in Table 2 of the [paper](https://arxiv.org/abs/2101.02815) will be stored in `output_dir/results_.json` and `output_dir/results_.txt` files.

## Parameters Description

Under Development

## Contact

For any queries related to library versions, datasets, script, and results please contact us here:

Email: [email protected]

Whatsapp: +91 9043751980