https://github.com/mbari-org/ecoz2-whale

ECOZ2 applied on whale song unit files
https://github.com/mbari-org/ecoz2-whale

hmm k-means linear-predictive-coding supervised-learning unsupervised-learning whale-song-files

Last synced: 7 months ago
JSON representation

ECOZ2 applied on whale song unit files

• Host: GitHub
• URL: https://github.com/mbari-org/ecoz2-whale
• Owner: mbari-org
• Created: 2018-09-27T00:17:07.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2020-11-16T17:03:51.000Z (almost 5 years ago)
• Last Synced: 2025-02-04T21:43:19.328Z (8 months ago)
• Topics: hmm, k-means, linear-predictive-coding, supervised-learning, unsupervised-learning, whale-song-files
• Language: Shell
• Homepage:
• Size: 230 MB
• Stars: 0
• Watchers: 4
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: ChangeLog.md

Awesome Lists containing this project

README

          
This space captures VQ/HMM classification exercises

on labelled whale song unit data.

## Status

Complete exercises with some level of model tuning.

A general presentation about VQ/HMM and its application on the

initial set of exercises described here can be found at

https://github.com/ecoz2/ecoz2-doc/blob/master/exploration/vqhmm-whale.pdf.

## Exercises

- [exerc00](exerc00): On 10 labelled files.

- [exerc01](exerc01): On a 4.5 hour recording.

- [exerc02](exerc02): As exerc01 but with some underlying adjustments in

  programs parameters and file organization.

- exerc03: related with the "windy" interval in the 4.5 recording:

  - [exerc03a](exerc03a): excluding the windy interval

  - [exerc03b](exerc03b): only considering the windy interval

- [exerc04](exerc04): As exerc01 but with number of refinements iterations

  set to 120

- exerc05: As exerc01 but with order of prediction P = 20

  according to rule of thumb P = 4 + fs / 1000,

  where the sampling frequency fs is 16kHz in our case.

  - [exerc05](exerc05a): Other parameters as in previous exercises

  - [exerc05b](exerc05b): Only considering classes with at least 100 instances

  - [exerc05c](exerc05c): Only considering classes with at least 200 instances

- exerc06: As exerc05c but merging classes "I" and "I2" into a "II" class.

  - [exerc06](exerc06)

## Some signal inspection

See [signal-inspection](signal-inspection).