-
Notifications
You must be signed in to change notification settings - Fork 11
/
checkinnov.py
158 lines (126 loc) · 5.34 KB
/
checkinnov.py
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
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
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
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
####
# KFTS quality check
# Script to display innovation and check for patterns
#
# Manon Dalaison 2020
####
#Global import
import numpy as np
import matplotlib
matplotlib.use('Agg') #so that don't ask for display!
import matplotlib.pyplot as plt
import h5py
import os
import datetime as dt
import configparser
import argparse
from ast import literal_eval
#Local import
import kf.readinput as infmt
from kf.timefunction import TimeFct
#----------------------------------------------------------------
#First get file name from inline argument
parser = argparse.ArgumentParser( description='Plot KFTS info in Updates.h5')
parser.add_argument('-c', type=str, dest='config', default=None,
help='Specify INI config file of KFTS')
args = parser.parse_args()
#Read config file
config = configparser.ConfigParser(interpolation=configparser.ExtendedInterpolation())
config.read(args.config)
loc = config['INPUT'].get('workdir', fallback='./')
outdir = os.path.join(loc, config['OUTPUT'].get('outdir', fallback=''))
locfig = os.path.join(loc, config['OUTPUT'].get('figdir', fallback=''))
datfile = os.path.join(loc, config['INPUT'].get('infile'))
fmtfile = config['INPUT'].get('fmtfile', fallback='ISCE')
model = literal_eval(config['MODEL SETUP'].get('model'))
EQ = config['MODEL SETUP'].getboolean('EQ', fallback=False)
eqinfo = config['INPUT'].get('eqinfo', fallback=None)
if eqinfo is not None:
eqinfo = os.path.join(loc,eqinfo)
# Adjust model in the case of earthquakes
if EQ == True:
#Get earthquake properties in actual reference frame
#file obtained by running "earthquake2step.py"
Xeq,Yeq,teq,Rinf,Aeq,sx,sy = np.loadtxt(eqinfo, unpack=True)
Xeq = Xeq.astype('int32')
Yeq = Yeq.astype('int32')
if isinstance(teq,float):
#specific case of one earthquake only
Leq = 1
teq = [teq]
Xeq,Yeq = [Xeq],[Yeq]
Rinf,sx,sy = [Rinf],np.array(sx),np.array(sy)
elif isinstance(teq,np.ndarray):
teq = teq.tolist()
Leq = len(teq)
else:
assert False, "Verify content of {}".format(eqinfo)
teq.insert(0,'STEP')
model.insert(len(model),tuple(teq))
print('New functional model with earthquakes :',model)
infile = os.path.join(outdir,'Updates.h5')
#---------------------------------------------------------------------
#Import
data = infmt.SetupKF(datfile, fmt=fmtfile)
tfct = TimeFct(data.time,model)
dates = [dt.datetime.fromordinal(dd) for dd in data.orddates]
#Import and read innovation
fin = h5py.File(infile,'r') #dictionary
innov = fin['mean_innov']
gain = fin['param_gain']
Ny, Nx, Nt, L = np.shape(gain)
labels = tfct.get_label(L,'mm')
#--------------------------------------------------------------------
#Compute arrays
meanI = np.nanmean(innov,axis=2)
medI = np.nanmedian(innov,axis=2)
stdI = np.nanstd(innov,axis=2)
lastI = innov[:,:,-1]
cin1,cax1 = np.median(meanI[meanI!=0.0]) +np.quantile(meanI[meanI!=0.0],0.999), \
np.median(meanI[meanI!=0.0]) -np.quantile(meanI[meanI!=0.0],0.999)
cin2,cax2 = np.median(medI[medI!=0.0]) +np.quantile(medI[medI!=0.0],0.999), \
np.median(medI[medI!=0.0]) -np.quantile(medI[medI!=0.0],0.999)
cin3,cax3 = 0., 3*np.quantile(stdI[stdI!=0.0],0.9)
cin4,cax4 = np.median(lastI[~np.isnan(lastI)&(lastI!=0.0)]) \
+np.quantile(lastI[~np.isnan(lastI)&(lastI!=0.0)],0.999), \
np.median(lastI[~np.isnan(lastI)&(lastI!=0.0)]) \
-np.quantile(lastI[~np.isnan(lastI)&(lastI!=0.0)],0.999)
if np.nanmean(innov,axis=2).max()/cax2 > 10**2:
print("WARNING: some pixels have extremely high innovation")
#--------------------------------------------------------------------
print("** Start plot **")
print("Mean", np.nanmean(innov,axis=2).min(),np.nanmean(innov,axis=2).max())
print("STD", np.nanstd(innov,axis=2).min(),np.nanstd(innov,axis=2).max())
print("last innov", innov[:,:,-1].min(),innov[:,:,-1].max())
#Plot mean median and std innov maps
fig,ax = plt.subplots(1,4,figsize=(13,4),sharex=True,sharey=True)
img0 = ax[0].imshow(meanI, vmin=cin1, vmax=cax1)
img1 = ax[1].imshow(medI, vmin=cin2, vmax=cax2)
img2 = ax[2].imshow(stdI, vmin=cin3, vmax=cax3)
img3 = ax[3].imshow(lastI, vmin=cin4, vmax=cax4)
plt.colorbar(img0,ax=ax[0],orientation='horizontal',shrink=0.7,aspect=12.)
plt.colorbar(img1,ax=ax[1],orientation='horizontal',shrink=0.7,aspect=12.)
plt.colorbar(img2,ax=ax[2],orientation='horizontal',shrink=0.7,aspect=12.)
plt.colorbar(img3,ax=ax[3],orientation='horizontal',shrink=0.7,aspect=12.)
ax[0].set_title("Mean innovation for all timesteps")
ax[1].set_title("Median innovation")
ax[2].set_title("Standard deviation of innovation")
ax[3].set_title("Innovation at last time step")
#Plot evolution of gain for 4 random pixels
Npix = 5
fig1,ax1 = plt.subplots(1,L,figsize=(13,3.5))
sx,sy = np.random.randint(0,Nx-1,size=Npix),np.random.randint(0,Ny-1,size=Npix)
for k in range(L):
for j in range(Npix):
toplot = gain[sy[j],sx[j],:,k]
toplot[toplot==0]=np.nan
ax1[k].plot(dates[-Nt:],toplot,'.')
ax1[k].set_title(labels[k])
ax1[k].set_xlabel('Time')
#print("labels are", ax1[k].xaxis.get_majorticklabels())
plt.setp( ax1[k].xaxis.get_majorticklabels(), rotation=70 )
ax1[0].set_ylabel('Gain')
fig1.tight_layout()
print("** Save plot **")
fig.savefig(locfig+'innovation_stats.png',dpi=150)
fig1.savefig(locfig+'gain_randompixels.png',dpi=150)