Licence and potential addition to SeisBench

[yetinam]

Hi! I recently found your dataset through the SRL publication. Thanks for compiling this useful set of data! I'm opening this issue for two questions. First, I didn't find any information under which license the dataset has been published. Could you specify this? Second, I was wondering if you'd be interested in adding the dataset to SeisBench? I think it would be great to have it in there in an interoperable format with lots of other datasets. The data would need to be converted, but I'd be happy to assist you in that.

[chenyk1990]

not sure if you received it, on my sis it says file too large for delivery.

…

On Thu, Jan 25, 2024 at 8:48 AM Y. Chen ***@***.***> wrote: Hi, Thanks for reaching out. 1. I just added the LICENSE statement in the repository. It is under GNU v3 license. 2. Sure, we would be happy if you could include it in SeisBench and make it helpful to more users. At the same time, it is worth mentioning our EQCCT phase picker (https://github.com/chenyk1990/eqcct), which has been applied in production by TexNet in a real-time way. It is the cutting-edge picker we have developed in the past two years. I attached the EQCCT paper, FYI. If you're interested in including EQCCT in Seisbench, that would be great. We have a student who leveraged Seisbench and EQCCT for picking earlier last summer. If you have any questions, please let me know. Kind regards, Yangkang Yangkang Chen, Ph.D. Research Scientist/Professor Lead of AI/ML research Texas Seismological Network (TexNet) Bureau of Economic Geology Jackson School of Geosciences The University of Texas at Austin On Thu, Jan 25, 2024 at 2:55 AM yetinam ***@***.***> wrote: > > Hi! I recently found your dataset through the SRL publication. Thanks for compiling this useful set of data! I'm opening this issue for two questions. First, I didn't find any information under which license the dataset has been published. Could you specify this? Second, I was wondering if you'd be interested in adding the dataset to SeisBench? I think it would be great to have it in there in an interoperable format with lots of other datasets. The data would need to be converted, but I'd be happy to assist you in that. > > — > Reply to this email directly, view it on GitHub, or unsubscribe. > You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[yetinam]

Hi @chenyk1990 ,

thanks for the quick reply and for clarifying the license. I'll have a look at the dataset and see about converting. That's probably going to be rather straightforward but I might come back with technical questions.

Regarding EQCCT, I'd be happy to include it in SeisBench, but it's probably going to be a bit more work. All models in SeisBench are reimplemented in PyTorch and should usually also come with a set of pretrained weights. I've you'd be interested in writing a pull request for this, I'd be happy to review and merge it. Personally, I don't have the capacity to code further model additions.