I use building tool: 1 time setup: go to Data>Set Building size>Advanced>[Set all relavent parking tags there I want by default]

1. Trace outline of two rows with building tool(B)
2. trace middle lane with way tool(A).
3. select middle way, then split(alt+x)
4. trace each parking spot top to bottom while connecting to middle
5. Select the new ways and the middle node and split the ways (P)
6. Select each line one by one and split then(alt+x)

“After making sure these are really background images and not just unmapped buildings in OpenStreetMap, we can put the negative samples into our dataset with a corresponding all-background mask. Then run rs weights again, update the dataset config, and re-train.”

How exactly is this done? You obtain imagery that has 0 buildings in it? then you add the imagery to images/20/…. ? I’m very confused by this part

How did you prepare the subset with rs subset to set it up for 80/10/10?

If you really wanted to codify something: name is in English by default unless in Quebec, where it would be French.

And Scruss I was just pointing out how there are differences between FR and EN. French put the French explorer name first and English put the English name first

@Circeus We do have different names in French and English, you just have to know where to look. In Ottawa/Gatineau for example: EN: MacDonald-Cartier Bridge FR: Pont Cartier-MacDonald See the difference? Some French/English feuds never die…

Building import is complete with 33% validated currently, but the validation process also includes splitting terraces and tagging building types instead of building=yes

Currently on the Ottawa side there’s about 27000 buildings still marked as building=yes. On the Gatineau side there’s about 16000 buildings still marked as building=yes

;)

isnt the abrieviated streets in Canada already complete?

Canada should be mostly complete…those that don’t have wikidata tags dont have a wikidata page

Eassailles Keypad-Mapper pour collectionner des addresses