Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - bacetiner

Pages: [1] 2
1
Thank you for your patience! I just pushed a fix to the BRAILS backend that addresses this issue.

I plan to make a few other BRAILS improvements before the end of this week, so this fix may not make it to a minor release until early next week. If you would prefer not to wait until the minor release is out, please use the following syntax to update your BRAILS from the GitHub master: pip install git+https://github.com/NHERI-SimCenter/BRAILS

Once the minor release is out, you can use pip install BRAILS or pip install BRAILS -U as usual to install the latest fixes/improvements.

I hope this helps!

2
Thank you for catching this issue!

I was able to recreate this problem. I am working on a patch and will post a fix soon.

Please stay tuned!

3
Regional Hazard Simulation (R2D, rWhale) / Re: BRAILS Tool in R2D
« on: February 07, 2024, 04:49:25 PM »
Thank you for your response. I will try to recreate the issue and get back to you with a solution. Stay tuned!

4
Regional Hazard Simulation (R2D, rWhale) / Re: BRAILS Tool in R2D
« on: February 06, 2024, 05:48:58 PM »
Thank you for using the SimCenter Forum!

I am happy to look into this issue and get it sorted out for you. For this purpose, could you please provide the bounding box coordinates, number of buildings and seed number you defined for your building inventory?

5
Absolutely, happy to help.

If you have any other questions, please let me know. Thank you again for using SimCenter Forum!

6
Here is the CSV file I get from BRAILS backend. It contains the building attributes required for performing earthquake damage and loss calculations for 1,569 buildings in the Mission Beach, CA area.

A few things to note about this inventory:
1) It omits 126 (1695-1569=126) buildings in Mission Beach for which street-level imagery could not be obtained.
2) The occupancy type column contains data imputed from the National Structure Inventory through BRAILS, as the current version of the software cannot obtain building occupancy classes at HAZUS-level granularity
3) This inventory contains a 13-floor hotel building. The simulation results for this building should be evaluated carefully, given that the HAZUS methodology may not be sufficiently detailed for this type of building.

I hope this helps!

7
Thank you all for your interest in BRAILS. The latest version of BRAILS fully addresses this issue.

If you have any further questions or comments, please let me know.

8
There has been a slight change in the way BRAILS stores building imagery. It now saves street-level and aerial imagery in your working directory in the tmp/images/street and tmp/images/satellite folders, respectively.

I hope this helps!

9
Thank you for using the SimCenter Forum!

The issue with setting the random seed and number of buildings in the BRAILS widget was addressed in R2D version 4.0.0. This new version of the R2D is currently only available on GitHub, but the installation media should be out soon.

On another note, I am able to run this query using the BRAILS backend just okay. If you are familiar with Python programming, I may be able to help you get this inventory using the BRAILS backend. The backend has several additional options that, in some instances, allow for better inventory generation. For example, for your region, by changing the source of the footprints, I am able to get an inventory of 1695 buildings (instead of 119 buildings as obtained from OpenStreetMaps).

Please let me know how I can assist you better!


10
Apologies for the slow response; I saw this follow-up question just now!

You are seeing this error because there are a total of 11 invalid footprint geometries (in this case, multi polygons that are not nested) in the provided GeoJSON. Anytime FootprintHandler hits one of these geometries, it throws a fatal error. Please see attached images for five randomly selected samples from these invalid footprint geometries.

I pushed an update to FootprintHandler so that it successfully bypasses such footprint issues. Please let me know if you have any problems after updating your BRAILS setup to the latest.

On a side note, I see that this GeoJSON was generated using OSM. When I look at the current footprints for the 11 problematic locations in the provided GeoJSON, I currently see valid footprints. I wonder if this issue has to do with the way the footprints were extracted from OSM. If you can share more information on the specific process used to obtain these footprints, I might be able provide suggestions that may eliminate this type of footprint extraction issues.

11
Thank you for your post!

I am unable to reproduce this error. Can you please try re-installing BRAILS directly from the GitHub repo using the syntax pip install git+https://github.com/NHERI-SimCenter/BRAILS and see if the issue persists? I wonder if this is a bug that went unnoticed in the latest release (i.e., pip install BRAILS installs the latest release of BRAILS) but got addressed since then.

If you continue to experience this issue after installing directly from the GitHub repo, could you please share the GeoJSON file you are getting this error with? This way, I can take a deeper look into the problem.

I look forward to your response!

12
Thank you for using the SimCenter Forum!

I believe you are getting this error because of the way your bounding box input is defined. For bounding box entries, BRAILS expects the location parameter defined as a tuple in the format location=(lon1,lat1,lon2,lat2). Here the longitude and latitude values are the coordinates of the opposite vertices of a rectangular bounding box (at the moment, BRAILS only takes rectangular bounding box input).

For instance, one can get a random set of 100 buildings contained in the West Los Angeles bounding shown in the attachment by defining the bounding box based on the coordinates of its top left vertex (-118.4546,34.0111) and bottom right vertex (-118.3866,34.0436) via the syntax:

invGenerator = InventoryGenerator(location=(-118.4546,34.0111,-118.3866,34.0436),
                                  nbldgs=100, randomSelection=True,
                                  GoogleAPIKey="YOUR-API-KEY")

I hope this helps! Please let me know if you run into any other issues.

13
Hi Amir,

Thank you for using the SimCenter Forum, and apologies that this message escaped our attention until today.

Unfortunately, this is not possible through Google Maps APIs. Can you elaborate a little more on your target application? We are relatively close to releasing a module that can take snapshots of satellite images at different time instants, yet at a lower resolution, and I would like to hear if this would be helpful with your application.

Thank you

14
Hi Niko,

Thank you for using the SimCenter Forum, and apologies that this message escaped our attention until today.

Something about this image seems to trigger an error with the way FacadeParser computes the roof run. For easier troubleshooting, any chance you could share the latitude/longitude of the building location that results in this error?

Thank you

15
Hi Niko,

Thank you for using the SimCenter Forum, and apologies that this message escaped our attention until today.

A quick look at the error shows that the ImHandler has difficulty extracting the scale for the retrieved image for this particular building. I will need to look deeper into this issue to identify the potential causes.

I will get back to you with an update once this issue is resolved.

Pages: [1] 2