Latest Contributions by DavidPike

It seems like a big oversight not to have this exposed as a parameter. ... View more

If it's just being passed as a parameter, try = 'RESNET34' ... View more

From a purely python perspective it's not understanding your class name 'models'. Try pytorch.models.resnet34 or whatever module you imported. ... View more

Hi Tom, It may be some profile data remaining in your AppData folder (mine is C:\Users\DaveP\AppData). If you cant see this folder in your C:\ drive, it may be hidden, just go to 'view' and tick the 'show hidden items' box.  This is dependant on OS though. Before you delete or rename the Esri folders, try to install the software on a different profile to check this is causing the problem. Bonjour Tom, Il se peut que certaines données de profil restent dans votre dossier AppData (le mien est C:\Users\DaveP\AppData). Si vous ne pouvez pas voir ce dossier dans votre lecteur C:\, il peut être masqué, allez simplement dans 'voir' et cochez la case 'afficher les éléments cachés'. Cela dépend cependant du système d'exploitation. Avant de supprimer ou de renommer les dossiers Esri, essayez d'installer le logiciel sur un profil différent pour vérifier que cela cause le problème. ... View more

Hi Krissy, Maybe a rough-and-ready representation of the dorms, rooms and their rough position would actually be better and easier to interpret than a full-on CAD for each dorm? Much like the London Underground (tube) map. I think that could be knocked up very quickly. ... View more

For your select query I think OBJECTID should be enclosed in quotes. select = ' " ' + "OBJECTID" + ' " '  + " = " + str(row[0]) ... View more

Looking at the code it seems absolutely fine. Is the comparison messing up on different data types, e.g.  string and float? ... View more

I'm not sure your list comprehension is working in your cursor. Check your list is being populated  if not, replace with a noob append on fclist. ... View more

I would get a raster DEM and use that to calculate new areas in raster calculator. ... View more

You could create the base using contours. The height of that contour area would correspond to the layer they belong to. Eg everything over 0m would be your base, the next layer is everything over 100m, next layer everything over 200m.. I dont know if this is obvious sorry. ... View more

As Duncan says, 'sliver' is not the right terminology. Without seeing the rest of the data, I would suggest selecting you main land body and then using select by location to select all features intersecting and within it. Export your selection to a new feature, or switch selection and delete. ... View more

Your ODC factors in the network distance, the line is purely representative. The attributes in the table will be accurate to network paths. You may also consider closest facility and establishing routes for each village with scripting. ... View more

I think the joins or relates are messing things up. Right click on your feature in the table of contents and go to data->export and create a new feature. ... View more

I would split your raster into new rasters for every landcover type then run zonal statistics sum for each. It would probably be easy to just manually delete the grid cells you dont want. Otherwise you could turn the raster into polygon then select by location and reverse selection and delete. For part b I dont really understand. If you just want the most occurring value zonal stats can use the majority function on the raster. ... View more

I've never seen that before.  Can you add a new field as float and use field calculator to calculate the same values and retry on that field. What does it look like if you based one of those values from the total column into excel? ... View more