-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathstack_filter.py
executable file
·207 lines (188 loc) · 8.99 KB
/
stack_filter.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
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
#! /usr/bin/env python
import sys
import os
import re
from datetime import datetime
import numpy as np
from lib import iolib
from lib import malib
from lib import timelib
if len(sys.argv) != 3:
sys.exit('Must specify input stack and mode')
mode = None
#Set this to output the inverse of the specified filter
inv = False
#Set this to output both the pass and stop
both = False
stack_fn = sys.argv[1]
mode = sys.argv[2]
#Load stack
#s = malib.DEMStack(stack_fn=stack_fn, stats=True, trend=True, med=True)
s = malib.DEMStack(stack_fn=stack_fn, stats=False, trend=False, med=False, datestack=False)
#For bed
#s = malib.DEMStack(stack_fn=stack_fn, trend=False, stats=False, datestack=False, med=False)
#Pull out Dutrieux bed
#idx = [3,]
#Use default output filename
#Note: if out_stack_fn is not specified, a new stack fn will be computed with updated min/max dt and count
out_fn = None
#out_fn = os.path.splitext(s.stack_fn)[0]+'_filt.npz'
if mode == 'pig_pre_tiltcorr':
#Isolate DG (both stereo and mono, trans and nocorr) and altimetry
idx = np.array([bool(re.search('DG|LVIS|ATM', i)) for i in s.source])
#Limit to 2009-present, when dh/dt is ~constant
#LVIS on 10/20/2009 should be first
min_dt = datetime(2009,1,1)
max_dt = datetime(2016,1,1)
idx *= (s.date_list >= min_dt) & (s.date_list <= max_dt)
both = True
out_fn = os.path.splitext(s.stack_fn)[0]+'_filt_DG+LVIS+ATM_2009-2016.npz'
elif mode == 'remove_nocorr':
#Isoalate remaining nocorr after tiltcorr
#Source should have tiltcorr appended for all that were adjusted
idx = np.array([bool(re.search('nocorr$', i)) for i in s.source])
both = True
out_fn = os.path.splitext(s.stack_fn)[0]+'_filt_nocorr.npz'
elif mode == 'remove_mono':
#Isoalate remaining nocorr after tiltcorr
#Source should have tiltcorr appended for all that were adjusted
idx = np.array([bool(re.search('mono', i)) for i in s.source])
both = True
out_fn = os.path.splitext(s.stack_fn)[0]+'_filt_mono.npz'
elif mode == 'pig_shelf_dt':
min_dt = datetime(2008,1,1)
max_dt = datetime(2015,6,1)
idx = np.array((s.date_list >= min_dt) & (s.date_list <= max_dt))
both = True
out_fn = os.path.splitext(s.stack_fn)[0]+'_filt_20080101-20150601.npz'
elif mode == 'pig_gps_dt':
min_dt = datetime(2011,8,1)
max_dt = datetime(2014,4,1)
idx = np.array((s.date_list >= min_dt) & (s.date_list <= max_dt))
both = True
out_fn = os.path.splitext(s.stack_fn)[0]+'_filt_20110801-20140401.npz'
#Pull out grids that cover entire GPS area
elif mode == 'pig_gps_test':
#good_dt = [datetime(2011,12,31).date(), datetime(2012,10,23).date()]
mincount = 20000
good_dt = [datetime(2012,1,14).date(), datetime(2012,2,2).date(), datetime(2012,10,23).date(), datetime(2012,11,11).date(), datetime(2013,9,24).date()]
idx = np.array([d.date() in good_dt for d in s.date_list])
good = s.ma_stack[idx]
count = good.reshape(good.shape[0], good.shape[1]*good.shape[2]).count(axis=1)
idx[idx] *= (count > mincount)
both = False
out_fn = os.path.splitext(s.stack_fn)[0]+'_filt_gpstest.npz'
else:
#Sample index creation for subset
#idx = [bool(re.search('ATM|LVIS|GLAS|DG_stereo$|DG_mono$', i)) for i in s.source]
#idx = [bool(re.search('ATM|LVIS|GLAS|SPIRIT|DG_stereo$|DG_mono$', i)) for i in s.source]
#idx = [bool(re.search('ATM|LVIS|GLAS|SPIRIT', i)) for i in s.source]
#idx = [bool(re.search('DG_stereo$', i)) for i in s.source]
#idx = [bool(re.search('mono', i)) for i in s.source]
#idx = [bool(re.search('DG_stereo_nocorr|DG_mono', i)) for i in s.source]
#idx = [bool(re.search('tiltcorr|medcorr', i)) for i in s.source]
#idx = [bool(re.search('aspbias', i)) for i in s.source]
#idx = [bool(re.search('DG_stereo_nocorr|DG_mono_nocorr', i)) for i in s.source]
#idx = [bool(re.search('DG_stereo_nocorr|DG_mono_nocorr', i)) for i in s.source]
#idx = [bool(re.search('SPIRIT', i)) for i in s.source]
#idx = [bool(re.search('DG|SPIRIT|TDM|GLISTIN', i)) for i in s.source]
#idx = [bool(re.search('TSX', i)) for i in s.source]
#idx = [bool(re.search('DG_stereo$|LVIS|ATM', i)) for i in s.source]
#Late season waves
#late_2015 = [datetime(2015, 8, 22, 15, 43), datetime(2015, 8, 5, 16, 11), datetime(2015, 7, 31, 15, 15), datetime(2015, 7, 19, 17, 12), datetime(2015, 7, 5, 0, 41)]
#late_2014 = [datetime(2014, 8, 7, 15, 50), datetime(2014, 7, 4, 15, 46), datetime(2014, 6, 23, 15, 8)]
#late_2013 = [datetime(2013, 9, 29, 15, 10), datetime(2013, 9, 8, 14, 41), datetime(2013, 8, 22, 15, 8), datetime(2013, 7, 5, 1, 6), datetime(2013, 6, 17, 15, 42)]
#late_2012 = [datetime(2012, 10, 31, 20, 32), datetime(2012, 9, 1, 20, 24), datetime(2012, 8, 26, 15, 14), datetime(2012, 8, 16, 15, 47), datetime(2012, 7, 20, 16, 2)]
#late_2011 = [datetime(2011, 9, 27, 15, 31), datetime(2011, 9, 14, 15, 37), datetime(2011, 9, 10, 15, 31), datetime(2011, 8, 14, 16, 9), datetime(2011, 8, 6, 16, 23), datetime(2011, 8, 2, 20, 24), datetime(2011, 7, 30, 15, 20)]
#late_season_dt = late_2015 + late_2014 + late_2013 + late_2012 + late_2011
#idx = [d in late_season_dt for d in s.date_list]
"""
#Approximate min/max DEMs for each year
mm2015 = [datetime(2015, 8, 22, 15, 43), datetime(2015, 4, 25, 17, 7)]
mm2014 = [datetime(2014, 8, 7, 15, 50), datetime(2014, 4, 19, 15, 34)]
mm2013 = [datetime(2013, 9, 29, 15, 10), datetime(2013, 3, 17, 17, 8)]
mm2012 = [datetime(2012, 8, 26, 15, 14), datetime(2012, 4, 9, 15, 43)]
mm2011 = [datetime(2011, 9, 27, 15, 31), datetime(2011, 4, 5, 15, 47)]
#Second is ATM
#Note: these were datetime for full jak catchment, not lower40km
#mm2010 = [datetime(2010, 8, 17, 15, 15), datetime(2010, 5, 14, 12, 48)]
mm2010 = [datetime(2010, 8, 17, 15, 15), datetime(2010, 5, 15, 18, 52)]
#Second is Moller
mm2009 = [datetime(2009, 8, 6, 15, 14), datetime(2009, 5, 6, 0, 0)]
#Second is ATM
#mm2009 = [datetime(2009, 8, 6, 15, 14), datetime(2009, 4, 29, 8, 32)]
#mm2009 = [datetime(2009, 8, 6, 15, 14), datetime(2009, 4, 28, 15, 03)]
mm2008 = [datetime(2008, 8, 2, 15, 5), datetime(2008, 6, 4, 15, 27)]
#First is SPIRIT, second ATM
#mm2007 = [datetime(2007, 8, 4, 15, 9), datetime(2007, 5, 10, 14, 26)]
mm2007 = [datetime(2007, 8, 4, 15, 9), datetime(2007, 5, 10, 14, 22)]
#Both are ATM
#mm2007 = [datetime(2007, 9, 19, 18, 6), datetime(2007, 5, 10, 14, 26)]
#mm2007 = [datetime(2007, 9, 20, 17, 8), datetime(2007, 5, 10, 14, 22)]
#Just max ATM
#mm2006 = [datetime(2006, 5, 27, 19, 41),]
#mm2005 = [datetime(2005, 5, 15, 15, 27),]
#mm2003 = [datetime(2003, 5, 11, 18, 59),]
mm2006 = [datetime(2006, 5, 27, 11, 35),]
mm2005 = [datetime(2005, 5, 15, 9, 45),]
mm2003 = [datetime(2003, 5, 11, 13, 15),]
mm = mm2015 + mm2014 + mm2013 + mm2012 + mm2011 + mm2010 + mm2009 + mm2008 + mm2007 + mm2006 + mm2005 + mm2003
idx = [d in mm for d in s.date_list]
"""
#Pull out end of summer data
min_rel_dt = (6, 1)
max_rel_dt = (12, 31)
#Pull out spring data
#min_rel_dt = (1, 1)
#min_rel_dt = (3, 1)
#max_rel_dt = (5, 31)
#max_rel_dt = (6, 1)
#idx = timelib.rel_dt_list_idx(s.date_list, min_rel_dt, max_rel_dt)
"""
dt_list = [ \
(datetime(2002,6,1), datetime(2003,5,31)), \
(datetime(2003,6,1), datetime(2004,5,31)), \
(datetime(2004,6,1), datetime(2005,5,31)), \
(datetime(2005,6,1), datetime(2006,5,31)), \
(datetime(2006,6,1), datetime(2007,5,31)), \
(datetime(2007,6,1), datetime(2008,5,31)), \
(datetime(2008,6,1), datetime(2009,5,31)), \
(datetime(2009,6,1), datetime(2010,5,31)), \
(datetime(2010,6,1), datetime(2011,5,31)), \
(datetime(2011,6,1), datetime(2012,5,31)), \
(datetime(2012,6,1), datetime(2013,5,31)), \
(datetime(2013,6,1), datetime(2014,5,31)), \
(datetime(2014,6,1), datetime(2015,5,31))]
for min_dt,max_dt in dt_list:
idx = (s.date_list >= min_dt) & (s.date_list <= max_dt)
s_sub = malib.get_stack_subset(s, idx, copy=True, save=True)
"""
#min_dt = datetime(2007,6,1)
min_dt = datetime(2007,1,1)
#max_dt = s.date_list.compressed()[-1]
#min_dt = datetime(2000,1,1)
#min_dt = datetime(2009,6,1)
#max_dt = datetime(2010,6,1)
#min_dt = datetime(2008,6,1)
#min_dt = datetime(2012,6,1)
max_dt = datetime(2016,1,1)
#max_dt = datetime(2015,12,31)
#max_dt = datetime(2011,6,1)
#idx *= (s.date_list >= min_dt) & (s.date_list <= max_dt)
#Make sure idx is array
idx = np.array(idx)
#Invert
if both:
if np.any(idx):
s_sub = malib.get_stack_subset(s, idx, out_stack_fn=out_fn, copy=True, save=True)
if np.any(~idx):
if out_fn is not None:
out_fn = os.path.splitext(out_fn)[0]+'inv.npz'
s_sub = malib.get_stack_subset(s, ~idx, out_stack_fn=out_fn, copy=True, save=True)
else:
if inv:
idx = ~idx
if out_fn is not None:
out_fn = os.path.splitext(out_fn)[0]+'inv.npz'
if np.any(idx):
s_sub = malib.get_stack_subset(s, idx, out_stack_fn=out_fn, copy=True, save=True)