Pythonを用いた空間統計分析のための衛星データ（LST, NDVI）の処理2

import h5py
import pandas as pd
import tables
from pyhdf.SD import SD, SDC
import numpy as np
import pandas as pd
from scipy.interpolate import griddata
import seaborn as sns
import matplotlib.pyplot as plt

import os
import glob
import csv

import jismesh.utils as ju


pop_27City = pd.read_csv("./population_1kmmesh_XY/
1kmmesh_population_XY_main27.csv",header=0,engine="python")

## データをKEY_CODEに対して昇順にしておく
pop_27City = pop_27City.sort_values('KEY_CODE')
pop_27City = pop_27City.reset_index(drop=True)


# 対象期間
annual = [2012,2013,2014,2015]


## LSTの処理

concat_list = []

# 6月から9月までの平均値
season_all = pd.read_csv("./population_1kmmesh_MinDistance/
LST/ALL_6to9_AVE_LST_Distance.csv",header=0,engine="python")


# 'InputID'（8桁のメッシュコード）で昇順 → 重複する"InputID"を削除
season_all = season_all.sort_values('InputID')
season_all = season_all.reset_index(drop=True)
season_all = season_all[~season_all.duplicated(
subset=["InputID"])].reset_index(drop=True)

concat_list.append(season_all.iloc[:,1:])
concat_list[0] = concat_list[0].rename(columns={
'TargetID': 'LST_6to9_ave', 'Distance': 'Min_D[m]LST_6to9_ave'})


# 各年のデータの処理
for i in range(len(annual)):
    annual_alls = pd.read_csv("./population_1kmmesh_MinDistance/
LST/" + str(annual[i]) + "_ALLannual_AVE_Distance.csv",
header=0,engine="python")
    
    # 'InputID'（8桁のメッシュコード）で昇順 → 重複削除
    annual_alls = annual_alls.sort_values('InputID')
    annual_alls = annual_alls.reset_index(drop=True)
    annual_alls = annual_alls[~annual_alls.duplicated(
    subset=["InputID"])].reset_index(drop=True)
    
    concat_list.append(annual_alls.iloc[:,1:])
    concat_list[i+1] = concat_list[i+1].rename(
    columns={'TargetID': 'LST_' + str(annual[i]) + 'annual_AVE', 
    'Distance': 'Min_D[m]LST_' + str(annual[i]) + 'annual_AVE'})


# 結合
concat1 = pd.concat([concat_list[0], concat_list[1],
concat_list[2],concat_list[3],concat_list[4]], axis=1)


## NDVIの処理
concat2_list = []


# 6月から9月までの平均値
season_all = pd.read_csv("./population_1kmmesh_MinDistance/
NDVI/ALL_6to9_AVE_NDVI_Distance.csv",header=0,engine="python")


# 'InputID'（8桁のメッシュコード）で昇順 → 重複する"InputID"を削除
season_all = season_all.sort_values('InputID')
season_all = season_all.reset_index(drop=True)
season_all = season_all[~season_all.duplicated(
subset=["InputID"])].reset_index(drop=True)

concat2_list.append(season_all.iloc[:,1:])
concat2_list[0] = concat_list[0].rename(columns={
'TargetID': 'NDVI_6to9_ave', 'Distance': 'Min_D[m]NDVI_6to9_ave'})

# 各年のデータの処理

for i in range(len(annual)):
    annual_alls = pd.read_csv(
    "./population_1kmmesh_MinDistance/NDVI/" + str(annual[i]) 
    + "_ALLannual_AVE_Distance.csv",header=0,engine="python")
    
    # 'InputID'（8桁のメッシュコード）で昇順 → 重複削除
    annual_alls = annual_alls.sort_values('InputID')
    annual_alls = annual_alls.reset_index(drop=True)
    annual_alls = annual_alls[~annual_alls.duplicated(
    subset=["InputID"])].reset_index(drop=True)
    
    concat2_list.append(annual_alls.iloc[:,1:])
    concat2_list[i+1] = concat2_list[i+1].rename(
    columns={'TargetID': 'NDVI_' + str(annual[i]) + 'annual_AVE', 
    'Distance': 'Min_D[m]NDVI_' + str(annual[i]) + 'annual_AVE'})

# 結合
concat2 = pd.concat([concat2_list[0], concat2_list[1],
concat2_list[2],concat2_list[3],concat2_list[4]], axis=1)


## 人口メッシュデータ＋「NDVIとLST」を更に結合

ALL_concat = pd.concat([pop_27City ,concat1, concat2], axis=1)

ALL_concat.to_csv("./population_1kmmesh_MinDistance/
populationConcat_NDVILST.csv",index = False)