Skip to content

xiaozw1994/DKN

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

DKN MTSC

Setup

Configure the experimental environment and tools listed below:

pip3 install -r require.txt

Dataset

Download UEA datasets

URL: http://timeseriesclassification.com/dataset.php

When obtained the UEA zip file, one should unzipthis file and add to:

repo:  ./data/

Implementation

Global Configuration

see: ./processing/data.py
####configure introduce:
####dataset repo
save_file =  "/mnt/54b93d63-868e-4b2c-aa09-c9109c3e67be/multivariateTSC/UCE"
#### GPU index used
GPU_number = 0
#### dataset name
dataset_name = "Beef"
###  coeffecient for loss function
loss_coefficient = 0.1
### a scalarable temporate for distillation
temperate = 1.0
#####  the weight of dual non-target class loss
beta = 1.0
##### the weight of dual target class loss
alpha = 1.0
###### learning rate
learning_rate = 0.0001
##### weight_decay value
weight_decay = 0.0005
#### training epoch
epoch = 200
#### save path
save_path =  "save_model/"

Run Example

python3 main.py -n 1 -b 1 -r 0.01 -s 50 -p 1 -e 500

About

Densely Knowledge-aware Network for Multivariate Time Series Classification

Resources

License

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages