-
Notifications
You must be signed in to change notification settings - Fork 107
/
base.py
510 lines (460 loc) · 22.2 KB
/
base.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
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
import abc
import os
import h5py
import numpy as np
import psutil
from scipy.sparse import csr_matrix
from buffalo.data import prepro
from buffalo.data.fileio import (chunking_into_bins,
sort_and_compressed_binarization)
from buffalo.misc import aux, log
class Data(object):
def __init__(self, opt, *args, **kwargs):
self.opt = aux.Option(opt)
self.tmp_root = opt.data.tmp_dir
if not os.path.isdir(self.tmp_root):
os.makedirs(self.tmp_root)
self.handle = None
self.header = None
self.prepro = prepro.PreProcess(self.opt.data)
self.value_prepro = self.prepro
if self.opt.data.value_prepro:
self.prepro = getattr(prepro, self.opt.data.value_prepro.name)(self.opt.data.value_prepro)
self.value_prepro = self.prepro
self.data_type = None
self.temp_file_list = []
@abc.abstractmethod
def create_database(self, filename, **kwargs):
pass
def show_info(self):
header = self.get_header()
vali_size = 0
if self.has_group("vali"):
g = self.get_group("vali")
vali_size = g.attrs["num_samples"]
info = "{name} Header({users}, {items}, {nnz}) Validation({vali} samples)"
info = info.format(name=self.name,
users=header["num_users"],
items=header["num_items"],
nnz=header["num_nnz"],
vali=vali_size)
return info
def open(self, data_path):
self.handle = h5py.File(data_path, "r")
self.path = data_path
self.verify()
def verify(self):
assert self.handle, "Database is not opened"
if self.get_header()["completed"] != 1:
raise RuntimeError("Database is corrupted or partially built. Please try again, after remove it.")
def get_header(self):
assert self.handle, "Database is not opened"
if not self.header:
self.header = {"num_nnz": self.handle.attrs["num_nnz"],
"num_users": self.handle.attrs["num_users"],
"num_items": self.handle.attrs["num_items"],
"completed": self.handle.attrs["completed"]}
return self.header
def get_scale_info(self, with_sppmi=False, chunk_size=100000):
ret = {k: self.handle.attrs[k] for k in ["num_users", "num_items", "num_nnz", "sppmi_nnz"]}
if with_sppmi:
ret["sppmi_nnz"] = self.handle.attrs["sppmi_nnz"]
db = self.handle["rowwise"]
num_nnz = ret["num_nnz"]
vsum = 0.0
for offset in range(0, num_nnz, chunk_size):
limit = min(num_nnz, offset + chunk_size)
vsum += np.sum(db["val"][offset: limit])
ret["vsum"] = vsum
return ret
def get_group(self, group_name="rowwise"):
assert group_name in ["rowwise", "colwise", "vali", "idmap", "sppmi"], "Unexpected group_name: {}".format(group_name)
assert self.handle, "DB is not opened"
group = self.handle[group_name]
return group
def has_group(self, name):
return name in self.handle
def _iterate_matrix(self, axis, use_repr_name, userids, itemids):
assert axis in ["rowwise", "colwise"], "Unexpected data axis: {}".format(axis)
assert self.handle, "Database is not opened"
group = self.handle[axis]
data_index = 0
for u, end in enumerate(group["indptr"]):
keys = group["key"][data_index:end]
vals = group["val"][data_index:end]
if use_repr_name:
for k, v in zip(keys, vals):
yield userids(u), itemids(k), v
else:
for k, v in zip(keys, vals):
yield u, k, v
data_index = end
def _iterate_stream(self, axis, use_repr_name, userids, itemids):
assert axis in ["rowwise"], "Unexpected date axis: {}".format(axis)
assert self.handle, "Database is not opened"
group = self.handle[axis]
data_index = 0
for u, end in enumerate(group["indptr"]):
keys = group["key"][data_index:end]
if use_repr_name:
for k in keys:
yield userids(u), itemids(k)
else:
for k in keys:
yield u, k
data_index = end
def iterate(self, axis="rowwise", use_repr_name=False) -> [int, int, float]:
"""Iterate over data
args:
group: which data group
use_repr_name: return representative name of internal ids
"""
userids, itemids = None, None
idmap = self.get_group("idmap")
if use_repr_name:
userids = lambda x: str(x)
if idmap["rows"].shape[0] != 0:
userids = lambda x: idmap["rows"][x].decode("utf-8", "ignore")
itemids = lambda x: str(x)
if idmap["cols"].shape[0] != 0:
itemids = lambda x: idmap["cols"][x].decode("utf-8", "ignore")
if axis == "colwise":
userids, itemids = itemids, userids
if self.opt.data.internal_data_type == "matrix":
return self._iterate_matrix(axis, use_repr_name, userids, itemids)
elif self.opt.data.internal_data_type == "stream":
return self._iterate_stream(axis, use_repr_name, userids, itemids)
def get(self, index, axis="rowwise") -> [int, int, float]:
if self.opt.data.internal_data_type == "matrix":
assert axis in ["rowwise", "colwise"], "Unexpected data axis: {}".format(axis)
assert self.handle, "Database is not opened"
group = self.handle[axis]
begin = 0 if index == 0 else group["indptr"][index - 1]
end = group["indptr"][index]
keys = group["key"][begin:end]
vals = group["val"][begin:end]
return (keys, vals)
elif self.opt.data.internal_data_type == "stream":
assert axis in ["rowwise"], "Unexpected date axis: {}".format(axis)
assert self.handle, "Database is not opened"
group = self.handle[axis]
begin = 0 if index == 0 else group["indptr"][index - 1]
end = group["indptr"][index]
keys = group["key"][begin:end]
return (keys,)
def close(self):
if self.handle:
self.handle.close()
self.handle = None
self.header = None
def temp_file_clear(self):
for path in self.temp_file_list:
if os.path.isfile(path):
os.remove(path)
self.temp_file_list = []
def _create_database(self, path, **kwargs):
# Create database structure
if os.path.exists(path):
self.logger.info(f"File {path} exists. To build new database, existing file {path} will be deleted.")
os.remove(path)
f = h5py.File(path, "w")
self.path = path
num_users, num_items, num_nnz = kwargs["num_users"], kwargs["num_items"], kwargs["num_nnz"]
f.create_group("rowwise")
f.create_group("colwise")
chunk_size = (min(1024 ** 3 - 1, num_nnz),)
for g in [f["rowwise"], f["colwise"]]:
g.create_dataset("key", (num_nnz,), dtype="int32", maxshape=(num_nnz,), chunks=chunk_size)
g.create_dataset("val", (num_nnz,), dtype="float32", maxshape=(num_nnz,), chunks=chunk_size)
f["rowwise"].create_dataset("indptr", (num_users,), dtype="int64", maxshape=(num_users,))
f["colwise"].create_dataset("indptr", (num_items,), dtype="int64", maxshape=(num_items,))
num_nnz = self._create_validation(f, **kwargs)
f.attrs["num_users"] = num_users
f.attrs["num_items"] = num_items
f.attrs["num_nnz"] = num_nnz
f.attrs["completed"] = 0
iid_max_col = kwargs["iid_max_col"]
uid_max_col = kwargs["uid_max_col"]
idmap = f.create_group("idmap")
idmap.create_dataset("rows", (num_users,), dtype=h5py.string_dtype("utf-8", length=uid_max_col),
maxshape=(num_users,))
idmap.create_dataset("cols", (num_items,), dtype=h5py.string_dtype("utf-8", length=iid_max_col),
maxshape=(num_items,))
return f
def _create_validation(self, f, **kwargs):
if not self.opt.data.validation:
return kwargs["num_nnz"]
_, num_nnz = kwargs["num_users"], kwargs["num_nnz"]
method = self.opt.data.validation.name
f.create_group("vali")
g = f["vali"]
g.attrs["method"] = method
g.attrs["n"] = 0
if method == "sample":
sz = min(self.opt.data.validation.max_samples,
int(num_nnz * self.opt.data.validation.p))
g.create_dataset("indexes", (sz,), dtype="int64", maxshape=(sz,))
# Watch out, create_working_data cannot deal with last line of data
# for validation data thus we have to reduce index 1.
g["indexes"][:] = np.random.choice(num_nnz - 1, sz, replace=False)
num_nnz -= sz
elif method in ["newest"]:
sz = kwargs["num_validation_samples"]
g.attrs["n"] = self.opt.data.validation.n
# We don"t need to reduce sample size for validation samples. It
# already applied on caller side.
g.create_dataset("row", (sz,), dtype="int32", maxshape=(sz,))
g.create_dataset("col", (sz,), dtype="int32", maxshape=(sz,))
g.create_dataset("val", (sz,), dtype="float32", maxshape=(sz,))
g.attrs["num_samples"] = sz
return num_nnz
def fill_validation_data(self, db, validation_data):
if not validation_data:
return
validation_data = [line.strip().split() for line in validation_data]
assert len(validation_data) == db["vali"].attrs["num_samples"], "Mismatched validation data"
num_users, num_items = db.attrs["num_users"], db.attrs["num_items"]
row = [int(r) - 1 for r, _, _ in validation_data] # 0-based
col = [int(c) - 1 for _, c, _ in validation_data] # 0-based
val = np.array([float(v) for _, _, v in validation_data], dtype=np.float32)
temp_mat = csr_matrix((val, (row, col)), (num_users, num_items))
db["vali"]["row"][:] = row
db["vali"]["col"][:] = col
db["vali"]["val"][:] = self.value_prepro(temp_mat.data)
def _prepare_validation_data(self):
if hasattr(self, "vali_data"):
return True
db = self.handle
num_users, num_items = db.attrs["num_users"], db.attrs["num_items"]
row = db["vali"]["row"][::]
col = db["vali"]["col"][::]
val = db["vali"]["val"][::]
_temp_mat = csr_matrix((val, (row, col)), (num_users, num_items))
indptr = _temp_mat.indptr[1:]
key = _temp_mat.indices
vali_rows = np.arange(len(indptr))[np.ediff1d(indptr, to_begin=indptr[0]) > 0]
vali_gt = {
u: set(key[indptr[u - 1]:indptr[u]]) if u != 0 else set(key[:indptr[0]])
for u in vali_rows}
validation_seen = {}
max_seen_size = 0
for rowid in vali_rows:
seen, *_ = self.get(rowid)
validation_seen[rowid] = set(seen)
max_seen_size = max(len(seen), max_seen_size)
validation_seen = validation_seen
validation_max_seen_size = max_seen_size
self.vali_data = {
"row": row,
"col": col,
"val": val,
"vali_rows": vali_rows,
"vali_gt": vali_gt,
"validation_seen": validation_seen,
"validation_max_seen_size": validation_max_seen_size
}
return True
def _sort_and_compressed_binarization(self, mm_path, num_lines, max_key, sort_key):
num_workers = psutil.cpu_count()
merged_bin = sort_and_compressed_binarization(
mm_path,
self.tmp_root,
num_lines, max_key, sort_key, num_workers)
return merged_bin
def _load_compressed_triplet_bin(self, db, job_files, num_lines, max_key, is_colwise=0):
self.logger.info("Load triplet files. Total job files: %s" % len(job_files))
INDPTR_SIZE = 8
RECORD_SIZE = 8
record_estimated_size = num_lines * RECORD_SIZE
indptr_estimated_size = max_key * INDPTR_SIZE
indptr_file = job_files[0]
job_files = job_files[1:]
with open(indptr_file, "rb") as fin:
indptr_total_size = fin.seek(0, 2)
fin.seek(0, 0)
assert indptr_estimated_size == indptr_total_size, f"Not valid indptr file size {indptr_total_size} (excepted: {indptr_estimated_size})"
indptr = np.frombuffer(fin.read(max_key * 8),
dtype=np.int64,
count=max_key)
db["indptr"][:max_key] = indptr
record_total_size = 0
data_index = 0
for job in job_files:
with open(job, "rb") as fin:
total_size = fin.seek(0, 2)
if total_size == 0:
continue
record_total_size += total_size
total_records = int(total_size / RECORD_SIZE)
fin.seek(0, 0)
data = np.frombuffer(fin.read(),
dtype=np.dtype([("i", "i"),
("v", "f")]),
count=total_records)
I, V = data["i"], data["v"]
if self.opt.data.value_prepro:
V = self.value_prepro(V.copy())
db["key"][data_index:data_index + total_records] = I
db["val"][data_index:data_index + total_records] = V
data_index += total_records
assert record_estimated_size == record_total_size, f"Not valid record file size {record_total_size} (excepted: {record_estimated_size})"
os.remove(indptr_file)
for path in job_files:
os.remove(path)
def _chunking_into_bins(self, mm_path, num_lines, max_key, sep_idx):
num_workers = psutil.cpu_count()
while num_workers > 20:
num_workers = int(num_workers / 2)
num_chunks = num_workers * 2
self.logger.info(f"Dividing into {num_chunks} chunks...")
job_files = chunking_into_bins(mm_path,
self.tmp_root,
total_lines=num_lines,
num_chunks=num_chunks,
sep_idx=sep_idx,
num_workers=num_workers)
return job_files
def _build_compressed_triplets(self, db, job_files, num_lines, max_key, is_colwise=0):
self.logger.info("Total job files: %s" % len(job_files))
with log.ProgressBar(log.INFO, total=len(job_files), mininterval=10) as pbar:
indptr_index = 0
data_index = 0
RECORD_SIZE = 12
prev_key = 0
for job in job_files:
with open(job, "rb") as fin:
total_size = fin.seek(0, 2)
if total_size == 0:
continue
total_records = int(total_size / RECORD_SIZE)
fin.seek(0, 0)
data = np.frombuffer(fin.read(),
dtype=np.dtype([("u", "i"),
("i", "i"),
("v", "f")]),
count=total_records)
U, I, V = data["u"], data["i"], data["v"]
if is_colwise:
U, I = I, U
if self.opt.data.value_prepro:
V = self.value_prepro(V.copy())
self.logger.debug("minU: {}, maxU: {}".format(U[0], U[-1]))
assert data_index + total_records <= num_lines, "Requests data size(%s) exceed capacity(%s)" % (data_index + total_records, num_lines)
db["key"][data_index:data_index + total_records] = I
db["val"][data_index:data_index + total_records] = V
diff = U[1:] - U[:-1]
max_diff = np.amax(diff) if len(diff) else 0
indptr = [data_index for _ in range(U[0] - prev_key)]
for i in range(max_diff):
indptr += (np.where(diff > i)[0] + data_index + 1).tolist()
indptr.sort()
db["indptr"][indptr_index:indptr_index + len(indptr)] = indptr
assert indptr_index + len(indptr) <= max_key
data_index += total_records
indptr_index += len(indptr)
prev_key = U[-1]
pbar.update(1)
db["indptr"][indptr_index:] = data_index
for path in job_files:
os.remove(path)
def _build_data(self,
db,
working_data_path,
validation_data,
target_groups=["rowwise", "colwise"],
sort=True):
available_mb = psutil.virtual_memory().available / 1024 / 1024
approximated_data_mb = 0
with open(working_data_path, "rb") as fin:
fin.seek(0, 2)
approximated_data_mb = db.attrs["num_nnz"] * 3 * 4 / 1024 / 1024
buffer_mb = int(max(1024, available_mb * 0.75))
disk_based = self.opt.data.get("disk_based", False)
# for each sides
for group, sep_idx, max_key in [("rowwise", 0, db.attrs["num_users"]),
("colwise", 1, db.attrs["num_items"])]:
if group not in target_groups:
continue
self.logger.info(f"Building compressed triplets for {group}...")
self.logger.info("Preprocessing...")
self.prepro.pre(db)
if approximated_data_mb * 1.2 < available_mb and not disk_based:
self.logger.info("In-memory Compressing ...")
job_files = self._sort_and_compressed_binarization(
working_data_path,
db.attrs["num_nnz"],
max_key,
sort_key=sep_idx + 1 if sort else -1)
self._load_compressed_triplet_bin(
db[group], job_files,
num_lines=db.attrs["num_nnz"],
max_key=max_key,
is_colwise=sep_idx)
else:
self.logger.info("Disk-based Compressing...")
if sort:
aux.psort(working_data_path,
tmp_dir=self.opt.data.tmp_dir,
key=sep_idx + 1,
buffer_mb=buffer_mb)
job_files = self._chunking_into_bins(working_data_path,
db.attrs["num_nnz"],
max_key,
sep_idx=sep_idx)
self._build_compressed_triplets(db[group],
job_files,
num_lines=db.attrs["num_nnz"],
max_key=max_key,
is_colwise=sep_idx)
self.prepro.post(db[group])
if group == "rowwise":
self.fill_validation_data(db, validation_data)
self.logger.info("Finished")
class DataOption(object):
def is_valid_option(self, opt) -> bool:
"""General type/logic checking"""
assert hasattr(opt["data"], "disk_based"), "disk_based not defined on data"
assert isinstance(opt["data"]["disk_based"], bool), "invalid type for data.disk_based"
if "validation" in opt["data"]:
assert opt["data"]["validation"]["name"] in ["sample", "newest"], "Unknown validation.name."
if opt["data"]["validation"]["name"] == "sample":
assert hasattr(opt["data"]["validation"], "max_samples"), "max_samples not defined on data.validation."
assert isinstance(opt["data"]["validation"]["max_samples"], int), "invalid type for data.validation.max_samples"
assert hasattr(opt["data"]["validation"], "p"), "not defined on data.validation.p"
assert isinstance(opt["data"]["validation"]["p"], float), "invalid type for data.validation.p"
if opt["data"]["validation"]["name"] in ["newest"]:
assert hasattr(opt["data"]["validation"], "max_samples"), "max_samples not defined on data.validation."
assert isinstance(opt["data"]["validation"]["max_samples"], int), "invalid type for data.validation.max_samples"
assert hasattr(opt["data"]["validation"], "n"), "not defined on data.validation.n"
assert isinstance(opt["data"]["validation"]["n"], int), "invalid type for data.validation.n"
return True
class DataReader(object):
def __init__(self, opt):
self.opt = opt
self.temp_file_list = []
def get_main_path(self):
return self.opt.input.main
def get_uid_path(self):
return self.opt.input.uid
def get_iid_path(self):
return self.opt.input.iid
def _get_temporary_id_list_path(self, obj, name):
field_name = f"temp_{name}"
if hasattr(self, field_name):
return getattr(self, field_name)
tmp_path = aux.get_temporary_file(self.opt.data.tmp_dir)
self.temp_file_list.append(tmp_path)
with open(tmp_path, "w") as fout:
if isinstance(obj, np.ndarray,) and obj.ndim == 1:
fout.write("\n".join(map(str, obj.tolist())))
elif isinstance(obj, (list,)):
fout.write("\n".join(map(str, obj)))
else:
raise RuntimeError(f"Unexpected data type for id list: {type(obj)}")
setattr(self, field_name, tmp_path)
return tmp_path
def temp_file_clear(self):
for path in self.temp_file_list:
if os.path.isfile(path):
os.remove(path)
self.temp_file_list = []