python - Pythonic way to get closest point for each point in a data frame (Nearest Neighbour) -
I have a list of places and towers I am trying to figure out, for each location, what is the nearest tower I work in a way that works, but I am pretty sure that it is a very inefficient way of doing it.
How do I do this more dragonically?
I have over 4,000 locations and 11,000 towers
- From the closest to the nearest 'tower' data frame reaching the frame.
Here's the code:
Nikttmtm = pd.DataFrame () for I, ( "Late", "Long"]] in Sthan_rojh: Tovr_kold [ "Duri_kmi"] = Tovr_clons.apli (lambda row: Duri_on_anit_sferr (Sthan_sresht [ "kick"], Sthan_ro [ "Long"], row [ 'DIGITAL_LATITUDE'], row [ 'DIGITAL_LONGITUDE']) * 6373, axis = 1) a = tower_coords.sort ([Distance_km '], ascending = 1) [: 1] [[ "SITE_NUMBER [" service _ Long "] [location =" ""] "Sthan_kang", "Sthan_kam"] = "position _ name "Long"] Nearest = The closest powder. Append (A) Print (i)
Tovr_kold looks like this:
SITE_NUMBER DIGITAL_LATITUDE DIGITAL_LONGITUDE 1 67.21 -30.432 ..
< First, use
min () to find the closest find you provide a
key after the repetition of arguments = "text =" post-text "itemprop =" text ">
can (which will be your lambda function).
it would be more dragon to use a class instead of like structure instead of hash your
a object of the PAP's implementation hash Like reach With the use of square properties are more efficient.
This entire work can be done in a
map operation. You are taking a structure and mapping the elements between 1-1 correspondence in any other structure, which makes it a
map .
In the end (and it's not about being phenolic but it's important), you should use a structure to find the nearest point.
Comments
Post a Comment