要准备
pandasgeopandas
具体实现
思想是这样的
将要转化的csv文件放到List里使用for循环遍历在for循环内
使用pandas读入csv选取需要进行计算的列赋值给csvFileCal将NaN设置为0(这是我此处的计算需要)再将算出的结果添加到csvFile末尾,这样既保留了NaN的值,又计算出了正确的结果(因为在计算中如果具有NaN,计算结果也会是NaN)将DataFrame转换为GeoDataFrame(这里仅有点对象,如果你还包含面对象的话还需要考虑别的方法)设定好坐标系为WGS84(espg:4326)调用GeoPandas的方法输出为shapefile文件
import pandas as pd
import geopandas as gpd
# The code below is used to convert the csv file into a shapefile.
nameList = ['india&Pakistan91_95', 'india&Pakistan96_00', 'india&Pakistan01_05', 'india&Pakistan06_10', 'india&Pakistan11_15', 'india&Pakistan16_20']
for name in nameList:
csvpath = r"D:\Desktop\StuInnovate\data\GTD\筛选\{}.csv".format(name)
csvFile = pd.read_csv(csvpath)
csvFileCal = csvFile[['latitude', 'longitude','nkill','nwound','property','nhostkid']]
csvFileCal.fillna(0, inplace=True) #将NaN设置为0
csvFileCal['severeIndex'] = csvFileCal['nkill'] + csvFileCal['nwound'] + csvFileCal['property'] + csvFileCal['nhostkid']
csvFile['severeIndex'] = csvFileCal['severeIndex']
geoGDF = gpd.GeoDataFrame(csvFile, geometry=gpd.points_from_xy(csvFile.longitude, csvFile.latitude))
geoGDF.crs = {'init': 'epsg:4326'}
geoGDF.to_file(r"D:\Desktop\StuInnovate\data\GTD\shp\{}.shp".format(name), driver='ESRI Shapefile')
相关链接
发表评论