Event_Table.csv

[wycding]

Hi Yiru:
I have a quick question regarding the SHRP2 dataset. After downloading it, I couldn’t find the Event_Table.csv file in the extracted folders. Thanks a lot.

[Yiru-Jiao]

Hi @wycding,

This table was arranged under the folder ./RawData/SHRP2/FileToUse/InsightTables/Event_Table.csv and could be available only if you apply for the original datasets

• E. Sears, M. A. Perez, K. Dan, T. Shimamiya, T. Hashimoto, M. Kimura, S. Yamada, and T. Seo. A Study on the Factors That Affect the Occurrence of Crashes and Near-Crashes. Version V2. 2019. URL: https://doi.org/10.15787/VTT1/FQLUWZ
• C. K. Layman, M. A. Perez, T. Sugino, and J. Eggert. Research of Driver Assistant System. Version V3. 2019. URL: https://doi.org/10.15787/VTT1/DEDACT
  and put them under the directory ./RawData/SHRP2/.

Best regards,
Yiru

[wycding]

Hi Yiru
I have finished training the GSSM model and noticed that there are no .npz files within the SafetyCriticalTestSet folder.
I completed the dataset authorization process, but I was only given access to part of the attached documents. The file Event_Table.csv is still missing. Could you please advise whether I need to submit a new application via email to obtain this file? Alternatively, is Event_Table.csv not must, and are there any alternative ways I can use instead?

[Yiru-Jiao]

Hi @wycding ,

I'll answer your question one by one.
First, the .npz files are output by organise_events.py. They are processed for model inference and therefore not in the released dataset.
Second, as said in my previous reply, the table is raw data. The screenshot you provided seems to be the dataset https://doi.org/10.15787/VTT1/EFYEJR
I would advise you to have a look at the dumped data, as I described in README.md: "Resulting data such as trained models, loss logs, and evaluation results are compressed as two zip files of the ./PreparedData/ and ./ResultData/ folders. They can be downloaded from https://doi.org/10.4121/9caa1e6c-9abd-4e36-ae28-c9ea4542d940." I don't think Event_Table.csv is there, but I'm quite sure you'll find everything to repeat the experiments there. However, a kind note: the dumped data can be used for experiment reproduction only, and you'll need your DUL for new research and publication.
I hope this reply will be helpful.

Best regards,
Yiru