dataset reading trouble

30 views
Skip to first unread message

Nilah Ravi Nair

unread,
Jan 6, 2023, 3:36:16 PM1/6/23
to OpenPack Dataset & Challenges2022
Hi, 

I have been trying to use the data to organise it for sensor modality. However, I am stuck with this error. Could you please guide me on how to tackle it please?

  File "C:\Users\nilah\OneDrive\Documents\GitHub\openpack-torch\openpack_torch\data\datasets.py", line 107, in load_dataset

    ts_sess, x_sess = optk.data.load_imu(

  File "C:\Users\nilah\anaconda3\lib\site-packages\openpack_toolkit\data\dataloader.py", line 174, in load_imu

    assert delta.max() < th, (

AssertionError: max difference is 31 [ms], but difference smaller than th=30 is allowed.

 

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.

An exception has occurred, use %tb to see the full traceback.


Thanks and Regards

Nilah

Naoya Yoshimura

unread,
Jan 6, 2023, 9:46:51 PM1/6/23
to Nilah Ravi Nair, OpenPack Dataset & Challenges2022
Hello Nilah,

What kind of sensors and trainig data are you using?
Because of device error, some session has missing data points.
In that case, our package thorws AssertionException  because we cannot align the timestamps from the selected streams.

The fastest solution to this problem will be identify the sessions with errors and exclude it.
Otherwise please interpolate the missing timesteps in advance.

I registered this issue in the GitHub

Best regards,
Naoya Yoshimura

………………………………………………
Naoya Yoshimura

Hara Lab,
Graduate Graduate School of Information Science and Technology,
Osaka University
e-mail: yoshim...@gmail.com
………………………………………………


2023年1月7日(土) 5:36 Nilah Ravi Nair <nilahn...@gmail.com>:
--
You received this message because you are subscribed to the Google Groups "OpenPack Dataset & Challenges2022" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpack+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpack/2a1b5722-198a-4901-84f1-05a16b2d6059n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages