https://github.com/amirbroker/cupydtw

Use Cuda for Dynamic Time Warping
https://github.com/amirbroker/cupydtw

cuda dtw dynamic-time-warping python

Last synced: about 2 months ago
JSON representation

Use Cuda for Dynamic Time Warping

• Host: GitHub
• URL: https://github.com/amirbroker/cupydtw
• Owner: amirbroker
• Created: 2025-02-18T21:46:53.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2025-03-04T05:24:02.000Z (over 1 year ago)
• Last Synced: 2025-03-04T06:26:18.085Z (over 1 year ago)
• Topics: cuda, dtw, dynamic-time-warping, python
• Language: Cuda
• Homepage:
• Size: 637 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
I use cuDTW++ (Ultra-Fast Dynamic Time Warping on CUDA-enabled GPUs: https://github.com/asbschmidt/cuDTW) to calculate the DTW between a target time series and multiple sequences.

I use script crtDTWvrbs.py in folder config to modify two files in folder cuDTW to set up cuDTW++ to calculate DTW with any number of sequences(in data folder) of any length and a target(in data folder) of any length.

![TW](https://github.com/user-attachments/assets/a8254a35-6d92-4d06-a668-72a0053ad717)