Frequency issue with NGSIM dataset

[guoyage]

First of all thank you very much for your wonderful research content, I am a little confused about the preprocessing of NGSIM dataset. You use the .mat file in (https://github.com/nachiket92/conv-social-pooling) to convert it into a txt file, but as far as I know, the data frequency in the mat file is still not 10Hz, not 5Hz, And I didn't see the relevant downclocking operation in your cslstm_mat_to_text.py.zip file. At this time, if we still set the observation step size to 15 and the prediction step size to 25 in birds_eye.yaml, is it unreasonable? ? Are the real step sizes 30 and 50 respectively (assuming the sampling frequency is 10Hz). Very much looking forward to your reply, thank you.

[vinit-katariya]

We will look into it and get back to you.

[guoyage]

Thank you for your reply and hope to get your result soon.

[guoyage]

We will look into it and get back to you.

What are the results of your investigation? I am looking forward to your reply.

[vinit-katariya]

Thank you for bringing this to our attention.

You are correct. The results in the paper for NGSIM are for a 10 Hz sampling rate. We will post the results with a 5 Hz sampling rate for NGSIM in some time.

[guoyage]

Thank you for your reply. I have done a 5Hz study, but the result of prediction performance is poor, which may be caused by my improper data processing. We sincerely hope that you can provide more official and up-to-date results with data processing documents. We look forward to your correction.

[guoyage]

Thank you for bringing this to our attention.

You are correct. The results in the paper for NGSIM are for a 10 Hz sampling rate. We will post the results with a 5 Hz sampling rate for NGSIM in some time.

What is your latest result? Looking forward to your reply

[guoyage]

Thank you for bringing this to our attention.

You are correct. The results in the paper for NGSIM are for a 10 Hz sampling rate. We will post the results with a 5 Hz sampling rate for NGSIM in some time.

What is your latest result? Looking forward to your reply