-
Notifications
You must be signed in to change notification settings - Fork 0
/
find_pairs.py
383 lines (313 loc) · 12.5 KB
/
find_pairs.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
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
import argparse
import os
import logging
from functools import partial
from multiprocessing import Pool, cpu_count, set_start_method
from numba import jit # type: ignore
import numpy as np
import pandas as pd
from methods.common.luc import luc_matching_columns
REPEAT_MATCH_FINDING = 100
DEFAULT_DISTANCE = 10000000.0
DEBUG = False
DISTANCE_COLUMNS = [
"elevation", "slope", "access",
"cpc0_u", "cpc0_d",
"cpc5_u", "cpc5_d",
"cpc10_u", "cpc10_d"
]
HARD_COLUMN_COUNT = 5
logging.basicConfig(level=logging.INFO, format="%(asctime)s [%(levelname)s] %(message)s")
def find_match_iteration(
k_parquet_filename: str,
m_parquet_filename: str,
start_year: int,
output_folder: str,
idx_and_seed: tuple[int, int]
) -> None:
logging.info("Find match iteration %d of %d", idx_and_seed[0] + 1, REPEAT_MATCH_FINDING)
rng = np.random.default_rng(idx_and_seed[1])
logging.info("Loading K from %s", k_parquet_filename)
# Methodology 6.5.7: For a 10% sample of K
k_set = pd.read_parquet(k_parquet_filename)
k_subset = k_set.sample(
frac=0.1,
random_state=rng
).reset_index()
logging.info("Loading M from %s", m_parquet_filename)
m_set = pd.read_parquet(m_parquet_filename)
# get the column ids for DISTANCE_COLUMNS
thresholds_for_columns = np.array([
200.0, # Elev
2.5, # Slope
10.0, # Access
0.1, # CPCs
0.1, # CPCs
0.1, # CPCs
0.1, # CPCs
0.1, # CPCs
0.1, # CPCs
])
logging.info("Preparing s_set...")
m_dist_thresholded_df = m_set[DISTANCE_COLUMNS] / thresholds_for_columns
k_subset_dist_thresholded_df = k_subset[DISTANCE_COLUMNS] / thresholds_for_columns
# convert to float32 numpy arrays and make them contiguous for numba to vectorise
m_dist_thresholded = np.ascontiguousarray(m_dist_thresholded_df, dtype=np.float32)
k_subset_dist_thresholded = np.ascontiguousarray(k_subset_dist_thresholded_df, dtype=np.float32)
# LUC columns are all named with the year in, so calculate the column names
# for the years we are intested in
luc0, luc5, luc10 = luc_matching_columns(start_year)
# As well as all the LUC columns for later use
luc_columns = [x for x in m_set.columns if x.startswith('luc')]
hard_match_columns = ['country', 'ecoregion', luc10, luc5, luc0]
assert len(hard_match_columns) == HARD_COLUMN_COUNT
# similar to the above, make the hard match columns contiguous float32 numpy arrays
m_dist_hard = np.ascontiguousarray(m_set[hard_match_columns].to_numpy()).astype(np.int32)
k_subset_dist_hard = np.ascontiguousarray(k_subset[hard_match_columns].to_numpy()).astype(np.int32)
# Methodology 6.5.5: S should be 10 times the size of K, in order to achieve this for every
# pixel in the subsample (which is 10% the size of K) we select 100 pixels.
required = 100
logging.info("Running make_s_set_mask... required: %d", required)
starting_positions = rng.integers(0, int(m_dist_thresholded.shape[0]), int(k_subset_dist_thresholded.shape[0]))
s_set_mask_true, no_potentials = make_s_set_mask(
m_dist_thresholded,
k_subset_dist_thresholded,
m_dist_hard,
k_subset_dist_hard,
starting_positions,
required
)
logging.info("Done make_s_set_mask. s_set_mask.shape: %a", {s_set_mask_true.shape})
s_set = m_set[s_set_mask_true]
potentials = np.invert(no_potentials)
k_subset = k_subset[potentials]
logging.info("Finished preparing s_set. shape: %a", {s_set.shape})
# Notes:
# 1. Not all pixels may have matches
results = []
matchless = []
s_set_for_cov = s_set[DISTANCE_COLUMNS]
logging.info("Calculating covariance...")
covarience = np.cov(s_set_for_cov, rowvar=False)
logging.info("Calculating inverse covariance...")
invconv = np.linalg.inv(covarience).astype(np.float32)
# Match columns are luc10, luc5, luc0, "country" and "ecoregion"
s_set_match = s_set[hard_match_columns + DISTANCE_COLUMNS].to_numpy(dtype=np.float32)
# this is required so numba can vectorise the loop in greedy_match
s_set_match = np.ascontiguousarray(s_set_match)
# Now we do the same thing for k_subset
k_subset_match = k_subset[hard_match_columns + DISTANCE_COLUMNS].to_numpy(dtype=np.float32)
# this is required so numba can vectorise the loop in greedy_match
k_subset_match = np.ascontiguousarray(k_subset_match)
logging.info("Starting greedy matching... k_subset_match.shape: %s, s_set_match.shape: %s",
k_subset_match.shape, s_set_match.shape)
add_results, k_idx_matchless = greedy_match(
k_subset_match,
s_set_match,
invconv
)
logging.info("Finished greedy matching...")
logging.info("Starting storing matches...")
for result in add_results:
(k_idx, s_idx) = result
k_row = k_subset.iloc[k_idx]
match = s_set.iloc[s_idx]
if DEBUG:
for hard_match_column in hard_match_columns:
if k_row[hard_match_column] != match[hard_match_column]:
print(k_row)
print(match)
raise ValueError("Hard match inconsistency")
results.append(
[k_row.lat, k_row.lng] + [k_row[x] for x in luc_columns + DISTANCE_COLUMNS] + \
[match.lat, match.lng] + [match[x] for x in luc_columns + DISTANCE_COLUMNS]
)
logging.info("Finished storing matches...")
for k_idx in k_idx_matchless:
k_row = k_subset.iloc[k_idx]
matchless.append(k_row)
columns = ['k_lat', 'k_lng'] + \
[f'k_{x}' for x in luc_columns + DISTANCE_COLUMNS] + \
['s_lat', 's_lng'] + \
[f's_{x}' for x in luc_columns + DISTANCE_COLUMNS]
results_df = pd.DataFrame(results, columns=columns)
results_df.to_parquet(os.path.join(output_folder, f'{idx_and_seed[1]}.parquet'))
matchless_df = pd.DataFrame(matchless, columns=k_set.columns)
matchless_df.to_parquet(os.path.join(output_folder, f'{idx_and_seed[1]}_matchless.parquet'))
logging.info("Finished find match iteration")
@jit(nopython=True, fastmath=True, error_model="numpy")
def make_s_set_mask(
m_dist_thresholded: np.ndarray,
k_subset_dist_thresholded: np.ndarray,
m_dist_hard: np.ndarray,
k_subset_dist_hard: np.ndarray,
starting_positions: np.ndarray,
required: int
):
m_size = m_dist_thresholded.shape[0]
k_size = k_subset_dist_thresholded.shape[0]
s_include = np.zeros(m_size, dtype=np.bool_)
k_miss = np.zeros(k_size, dtype=np.bool_)
for k in range(k_size):
matches = 0
k_row = k_subset_dist_thresholded[k, :]
k_hard = k_subset_dist_hard[k]
for index in range(m_size):
m_index = (index + starting_positions[k]) % m_size
m_row = m_dist_thresholded[m_index, :]
m_hard = m_dist_hard[m_index]
should_include = True
# check that every element of m_hard matches k_hard
hard_equals = True
for j in range(m_hard.shape[0]):
if m_hard[j] != k_hard[j]:
hard_equals = False
if not hard_equals:
should_include = False
else:
for j in range(m_row.shape[0]):
if abs(m_row[j] - k_row[j]) > 1.0:
should_include = False
if should_include:
s_include[m_index] = True
matches += 1
# Don't find any more M's
if matches == required:
break
k_miss[k] = matches == 0
return s_include, k_miss
# Function which returns a boolean array indicating whether all values in a row are true
@jit(nopython=True, fastmath=True, error_model="numpy")
def rows_all_true(rows: np.ndarray):
# Don't use np.all because not supported by numba
# Create an array of booleans for rows in s still available
all_true = np.ones((rows.shape[0],), dtype=np.bool_)
for row_idx in range(rows.shape[0]):
for col_idx in range(rows.shape[1]):
if not rows[row_idx, col_idx]:
all_true[row_idx] = False
break
return all_true
@jit(nopython=True, fastmath=True, error_model="numpy")
def greedy_match(
k_subset: np.ndarray,
s_set: np.ndarray,
invcov: np.ndarray
):
# Create an array of booleans for rows in s still available
s_available = np.ones((s_set.shape[0],), dtype=np.bool_)
total_available = s_set.shape[0]
results = []
matchless = []
s_tmp = np.zeros((s_set.shape[0],), dtype=np.float32)
for k_idx in range(k_subset.shape[0]):
k_row = k_subset[k_idx, :]
hard_matches = rows_all_true(s_set[:, :HARD_COLUMN_COUNT] == k_row[:HARD_COLUMN_COUNT]) & s_available
hard_matches = hard_matches.reshape(
-1,
)
if total_available > 0:
# Now calculate the distance between the k_row and all the hard matches we haven't already matched
s_tmp[hard_matches] = batch_mahalanobis_squared(
s_set[hard_matches, HARD_COLUMN_COUNT:], k_row[HARD_COLUMN_COUNT:], invcov
)
# Find the index of the minimum distance in s_tmp[hard_matches] but map it back to the index in s_set
if np.any(hard_matches):
min_dist_idx = np.argmin(s_tmp[hard_matches])
min_dist_idx = np.arange(s_tmp.shape[0])[hard_matches][min_dist_idx]
results.append((k_idx, min_dist_idx))
s_available[min_dist_idx] = False
total_available -= 1
else:
matchless.append(k_idx)
return (results, matchless)
# optimised batch implementation of mahalanobis distance that returns a distance per row
@jit(nopython=True, fastmath=True, error_model="numpy")
def batch_mahalanobis_squared(rows, vector, invcov):
# calculate the difference between the vector and each row (broadcasted)
diff = rows - vector
# calculate the distance for each row in one batch
dists = (np.dot(diff, invcov) * diff).sum(axis=1)
return dists
def find_pairs(
k_parquet_filename: str,
m_parquet_filename: str,
start_year: int,
seed: int,
output_folder: str,
processes_count: int
) -> None:
logging.info("Starting find pairs")
os.makedirs(output_folder, exist_ok=True)
rng = np.random.default_rng(seed)
iteration_seeds = zip(range(REPEAT_MATCH_FINDING), rng.integers(0, 1000000, REPEAT_MATCH_FINDING))
with Pool(processes=processes_count) as pool:
pool.map(
partial(
find_match_iteration,
k_parquet_filename,
m_parquet_filename,
start_year,
output_folder
),
iteration_seeds
)
def main():
# If you use the default multiprocess model then you risk deadlocks when logging (which we
# have hit). Spawn is the default on macOS, but not on Linux.
set_start_method("spawn")
parser = argparse.ArgumentParser(description="Takes K and S and finds 100 sets of matches.")
parser.add_argument(
"--k",
type=str,
required=True,
dest="k_filename",
help="Parquet file containing pixels from K as generated by calculate_k.py"
)
parser.add_argument(
"--m",
type=str,
required=True,
dest="m_filename",
help="Parquet file containing pixels from M as generated by build_m_table.py"
)
parser.add_argument(
"--start_year",
type=int,
required=True,
dest="start_year",
help="Year project started."
)
parser.add_argument(
"--seed",
type=int,
required=True,
dest="seed",
help="Random number seed, to ensure experiments are repeatable."
)
parser.add_argument(
"--output",
type=str,
required=True,
dest="output_directory_path",
help="Directory into which output matches will be written. Will be created if it does not exist."
)
parser.add_argument(
"-j",
type=int,
required=False,
default=round(cpu_count() / 2),
dest="processes_count",
help="Number of concurrent threads to use."
)
args = parser.parse_args()
find_pairs(
args.k_filename,
args.m_filename,
args.start_year,
args.seed,
args.output_directory_path,
args.processes_count
)
if __name__ == "__main__":
main()