search

tungk / OED

Outlier Detection for Time Series with Recurrent Autoencoder Ensembles
GNU General Public License v3.0
70 stars 13 forks source link

question about the ECG dataset #1

Open ResearcherLifeng opened 4 years ago

ResearcherLifeng commented 4 years ago

Hi! Dear authors,

I have a question about the ECG dataset you used in your experiments.

I downloaded the ECG dataset from http://www.cs.ucr.edu/~eamonn/discords/ECG_data.zip But it seems this dataset is without any labels (for anomalies). Could you provide the labels you use in your work? or tell me how can I get the label?

Thanks, Lifeng

JunShengB commented 4 years ago

Do you solve the problem？

jogrundy commented 3 years ago

There doesn't seem to be a reference for the ECG dataset in the paper, nor is there any detail on feature extraction (if any) or how anomalies are labelled. There does seem to be more than one ECG dataset. More detail would be helpful.

B-Seif commented 3 years ago

Dear author, Please, can you give us the real ECG Data you have used ? the link you provided in the paper give a mat file; however in your code you use a csv file (utilis.py). Thanks

tungk commented 3 years ago

Hi, I think the author of the dataset modified the dataset. I download it as zip file and inside the zip file is csv files. In addition, the dataset I used has label.

B-Seif commented 3 years ago

Thank you for your reply, really I appreciated this. Do you still have a version of this data somewhere? could you upload this please?

I am looking for multi-variate time series dataset with labels for each observation and your data corresponds exactly to what I am looking for.

Otherwise Can you confirm that the cnn_seq_seq.py file is your first paper please?

Tung Kieu, Bin Yang, and Christian S. Jensen. Outlier detection for multidimensional time series using deep neural networks. In MDM, pages 125–134, 2018

qingyuanxingsi commented 3 years ago

As above, could you please so kind to share your ECG dataset?