interpolate.py 3.69 KB
Newer Older
eudesyawog's avatar
eudesyawog committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
import os
import numpy as np
import pandas as pd
import scipy as sp
from scipy import sparse, linalg
import seaborn as sns
import matplotlib.pyplot as plt
sns.set()

def smooth_plot (csv,lstBands):
    
    ptrn = os.path.basename(csv)
    year = ptrn.split('_')[1]
    outFolder = "./interpolate"
    if not os.path.exists(outFolder):
        os.makedirs(outFolder)

    df = pd.read_csv(csv)

    outdf = None
    ggplot_dic = {}
    for band in lstBands:
        cols = [col for col in df.columns if col.startswith(band) and col.endswith("Mean")]
        cols.sort()
        
eudesyawog's avatar
eudesyawog committed
26
        times = pd.date_range(start="{}0502".format(year), end = "{}1030".format(year), freq='D')
eudesyawog's avatar
eudesyawog committed
27 28 29 30 31 32 33 34 35 36 37 38 39
        dic = {}

        for index, row in df.iterrows():
            lstValues = []
            for j in range(len(times)): # dates
                date = str(times.strftime('{}_%Y%m%d_Mean'.format(band))[j])
                if date in cols: 
                    lstValues.append(row[date])
                else :
                    lstValues.append(np.nan)
            upsampled = pd.Series(lstValues,index=times)
            interpolated = upsampled.interpolate(method='linear',limit_direction='both')
            resampled = interpolated.resample('5D').mean()
eudesyawog's avatar
eudesyawog committed
40
            
eudesyawog's avatar
eudesyawog committed
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
            dic.setdefault('feat_index',[]).append(index)
            if band == lstBands[0] :
                dic.setdefault('ID',[]).append(row['ID'])
                dic.setdefault('Biom_f',[]).append(row['Biom_f'])
                dic.setdefault('Biom_s',[]).append(row['Biom_s'])
                dic.setdefault('Rdt_f',[]).append(row['Rdt_f'])
                dic.setdefault('Rdt_s',[]).append(row['Rdt_s'])
            for idx , value in resampled.items():
                dic.setdefault('{}_{}'.format(band,idx.strftime("%Y%m%d")),[]).append(value)

                ggplot_dic.setdefault('Date',[]).append(idx.strftime("%Y-%m-%d"))
                ggplot_dic.setdefault('Band',[]).append(band)
                ggplot_dic.setdefault('Plot',[]).append(row['ID'])
                # ggplot_dic.setdefault('Crop',[]).append(row['Crop'])
                ggplot_dic.setdefault('Sowing',[]).append(row['Sowing'])
                ggplot_dic.setdefault('Harvest',[]).append(row['Harvest'])
                ggplot_dic.setdefault('Value',[]).append(value)
        
        if outdf is None :
            outdf = pd.DataFrame.from_dict(dic)
        else:
            outdf = outdf.merge(pd.DataFrame.from_dict(dic),on='feat_index',how='inner')

    outCSV = os.path.join(outFolder,'{}'.format(ptrn.replace(".csv","_interpolate.csv")))
    outdf = outdf.drop(columns='feat_index')
    outdf.to_csv(outCSV,index=False)

    # To ggplot
    plotCSV = os.path.join(outFolder,'{}'.format(ptrn.replace(".csv","_interpolate_toggplot.csv")))
    ggplot_df = pd.DataFrame.from_dict(ggplot_dic)
    ggplot_df = ggplot_df.loc[ggplot_df.Plot.isin(outdf['ID'])]
    ggplot_df.to_csv(plotCSV,index=False)

if __name__=='__main__':

    csv = "./stats/niakhar_2017_opt_gapf_notree.csv"
    smooth_plot(csv,["B2","B3","B4","B8","B5","B6","B7","B8A","B11","B12","NDVI","NDWI","EVI","MSAVI2","GDVI","CIGreen","CIRedEdge"])

    csv = "./stats/niakhar_2017_radar_notree.csv"
    smooth_plot(csv,["VH","VV"])

    csv = "./stats/niakhar_2018_opt_gapf_notree.csv"
    smooth_plot(csv,["B2","B3","B4","B8","B5","B6","B7","B8A","B11","B12","NDVI","NDWI","EVI","MSAVI2","GDVI","CIGreen","CIRedEdge"])

    csv = "./stats/niakhar_2018_radar_notree.csv"
    smooth_plot(csv,["VH","VV"])

    csv = "./stats/nioro_2018_opt_gapf_notree.csv"
    smooth_plot(csv,["B2","B3","B4","B8","B5","B6","B7","B8A","B11","B12","NDVI","NDWI","EVI","MSAVI2","GDVI","CIGreen","CIRedEdge"])

    csv = "./stats/nioro_2018_radar_notree.csv"
    smooth_plot(csv,["VH","VV"])