使用Pandas＆＃x27;在GeodataFrames上应用（）函数

发布于 2025-01-18 06:03:09 字数 701 浏览 4 评论 0原文

我正在尝试做一件非常简单的事情。我有一个 geopandas 地理数据框 (gdf)，其中包含一个属性列 (gdf.att) 和一个几何列 (gdf.geometry)，所有几何图形都是形状点）。

现在我想创建一个额外的列，其值是点坐标和属性的函数的结果。我尝试使用 apply() 来执行此操作，如下所示：

gdf['new_col'] = gdf.apply(lambda row : function([row.geometry.x, row.geometry.y], row.att))

但是，我收到以下错误：

AttributeError: 'Series' object has no attribute 'geometry'

似乎在 apply 函数中，行是一系列我无法调用几何图形的序列。但我不明白这一点，因为如果我运行下面的测试来检查如果我只返回 row 会得到什么，那么我会得到一个 GeoDataFrame 作为输出。

test = gdf.apply(lambda row: row) # after running this, test is the same as gdf

有人知道如何将 apply() 与使用多列的 GDF 一起使用吗？

谢谢！

原文

I am trying to do a very simple thing. I have a geopandas geodataframe (gdf) with an attribute column (gdf.att) and a geometry column (gdf.geometry, all geometries are shapely points).

Now I want to create an extra column whose values are the result of a function of the point coordinates and the attribute. I'm trying to do this using apply() as below:

gdf['new_col'] = gdf.apply(lambda row : function([row.geometry.x, row.geometry.y], row.att))

However, I get the following error:

AttributeError: 'Series' object has no attribute 'geometry'

It seems that within the apply function, row is a series of which I cannot call the geometry. But I don't understand this because if I run the test below to check what I get if I just return row, then I get a GeoDataFrame as output.

test = gdf.apply(lambda row: row) # after running this, test is the same as gdf

Does someone know how to use apply() with GDF's using multiple columns?

Thanks!

分享到QQ

分享到微博

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

需要登录才能够评论，你可以免费注册一个本站的账号。

∞梦里开花 2025-01-25 06:03:09

创建了与您的问题语句匹配的MWE，
您错过了axis = 1 https://pandas.pydata.org/docs/reference/reference/api/pandas.dataframe.apply.html ，因此您应用于每个列，而不是每行。

import geopandas as gpd

gdf = gpd.read_file(gpd.datasets.get_path('naturalearth_cities')).rename(columns={"name":"att"})

def function(point, att):
    return att[int(point[0] % len(att))]

gdf['new_col'] = gdf.apply(lambda row : function([row.geometry.x, row.geometry.y], row.att), axis=1)

gdf.head(5)

	att	几何	new_col
0	梵蒂冈城区	（12.453386544971766 41.90328217960115）	1
SAN	MARINO	（12.441770157800141 77429357）	O
Luxembourg	Point	Z	3
4	POINT	（6.130002806227083 49.611660379121076）	Point
V	Palikir	（158.1499743237623 6.9166643696007725）	k

created a MWE that matches your problem statement
you have missed axis=1 https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.apply.html, hence you are applying to each column, not each row.

import geopandas as gpd

gdf = gpd.read_file(gpd.datasets.get_path('naturalearth_cities')).rename(columns={"name":"att"})

def function(point, att):
    return att[int(point[0] % len(att))]

gdf['new_col'] = gdf.apply(lambda row : function([row.geometry.x, row.geometry.y], row.att), axis=1)

gdf.head(5)

	att	geometry	new_col
0	Vatican City	POINT (12.453386544971766 41.903282179960115)	V
1	San Marino	POINT (12.441770157800141 43.936095834768004)	n
2	Vaduz	POINT (9.516669472907267 47.13372377429357)	z
3	Luxembourg	POINT (6.130002806227083 49.611660379121076)	o
4	Palikir	POINT (158.1499743237623 6.916643696007725)	k