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
|
# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Tests for DecodeCSV op from parsing_ops."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import numpy as np
from tensorflow.python.eager import context
from tensorflow.python.framework import errors
from tensorflow.python.framework import test_util
from tensorflow.python.ops import parsing_ops
from tensorflow.python.platform import test
@test_util.run_all_in_graph_and_eager_modes
class DecodeCSVOpTest(test.TestCase):
def _test(self, args, expected_out=None, expected_err_re=None):
if expected_err_re is None:
decode = parsing_ops.decode_csv(**args)
out = self.evaluate(decode)
for i, field in enumerate(out):
if field.dtype == np.float32 or field.dtype == np.float64:
self.assertAllClose(field, expected_out[i])
else:
self.assertAllEqual(field, expected_out[i])
else:
with self.assertRaisesOpError(expected_err_re):
decode = parsing_ops.decode_csv(**args)
self.evaluate(decode)
def testSimple(self):
args = {
"records": ["1", "2", '"3"'],
"record_defaults": [[1]],
}
expected_out = [[1, 2, 3]]
self._test(args, expected_out)
def testSimpleWithScalarDefaults(self):
args = {
"records": ["1,4", "2,5", "3,6"],
"record_defaults": [1, 2],
}
expected_out = [[1, 2, 3], [4, 5, 6]]
self._test(args, expected_out)
def testSimpleWith2DDefaults(self):
args = {
"records": ["1", "2", "3"],
"record_defaults": [[[0]]],
}
if context.executing_eagerly():
err_spec = errors.InvalidArgumentError, (
"Each record default should be at "
"most rank 1.")
else:
err_spec = ValueError, "Shape must be at most rank 1 but is rank 2"
with self.assertRaisesWithPredicateMatch(*err_spec):
self._test(args)
def testSimpleNoQuoteDelimiter(self):
args = {
"records": ["1", "2", '"3"'],
"record_defaults": [[""]],
"use_quote_delim": False,
}
expected_out = [[b"1", b"2", b'"3"']]
self._test(args, expected_out)
def testScalar(self):
args = {"records": '1,""', "record_defaults": [[3], [4]]}
expected_out = [1, 4]
self._test(args, expected_out)
def test2D(self):
args = {"records": [["1", "2"], ['""', "4"]], "record_defaults": [[5]]}
expected_out = [[[1, 2], [5, 4]]]
self._test(args, expected_out)
def test2DNoQuoteDelimiter(self):
args = {
"records": [["1", "2"], ['""', '"']],
"record_defaults": [[""]],
"use_quote_delim": False
}
expected_out = [[[b"1", b"2"], [b'""', b'"']]]
self._test(args, expected_out)
def testDouble(self):
args = {
"records": ["1.0", "-1.79e+308", '"1.79e+308"'],
"record_defaults": [np.array([], dtype=np.double)],
}
expected_out = [[1.0, -1.79e+308, 1.79e+308]]
self._test(args, expected_out)
def testInt64(self):
args = {
"records": ["1", "2", '"2147483648"'],
"record_defaults": [np.array([], dtype=np.int64)],
}
expected_out = [[1, 2, 2147483648]]
self._test(args, expected_out)
def testComplexString(self):
args = {
"records": ['"1.0"', '"ab , c"', '"a\nbc"', '"ab""c"', " abc "],
"record_defaults": [["1"]]
}
expected_out = [[b"1.0", b"ab , c", b"a\nbc", b'ab"c', b" abc "]]
self._test(args, expected_out)
def testMultiRecords(self):
args = {
"records": ["1.0,4,aa", "0.2,5,bb", "3,6,cc"],
"record_defaults": [[1.0], [1], ["aa"]]
}
expected_out = [[1.0, 0.2, 3], [4, 5, 6], [b"aa", b"bb", b"cc"]]
self._test(args, expected_out)
def testNA(self):
args = {
"records": ["2.0,NA,aa", "NA,5,bb", "3,6,NA"],
"record_defaults": [[0.0], [0], [""]],
"na_value": "NA"
}
expected_out = [[2.0, 0.0, 3], [0, 5, 6], [b"aa", b"bb", b""]]
self._test(args, expected_out)
def testWithDefaults(self):
args = {
"records": [",1,", "0.2,3,bcd", "3.0,,"],
"record_defaults": [[1.0], [0], ["a"]]
}
expected_out = [[1.0, 0.2, 3.0], [1, 3, 0], [b"a", b"bcd", b"a"]]
self._test(args, expected_out)
def testWithDefaultsAndNoQuoteDelimiter(self):
args = {
"records": [",1,", "0.2,3,bcd", '3.0,,"'],
"record_defaults": [[1.0], [0], ["a"]],
"use_quote_delim": False,
}
expected_out = [[1.0, 0.2, 3.0], [1, 3, 0], [b"a", b"bcd", b"\""]]
self._test(args, expected_out)
def testWithTabDelim(self):
args = {
"records": ["1\t1", "0.2\t3", "3.0\t"],
"record_defaults": [[1.0], [0]],
"field_delim": "\t"
}
expected_out = [[1.0, 0.2, 3.0], [1, 3, 0]]
self._test(args, expected_out)
def testWithoutDefaultsError(self):
args = {
"records": [",1", "0.2,3", "3.0,"],
"record_defaults": [[1.0], np.array([], dtype=np.int32)]
}
self._test(
args, expected_err_re="Field 1 is required but missing in record 2!")
def testWrongFieldIntError(self):
args = {
"records": [",1", "0.2,234a", "3.0,2"],
"record_defaults": [[1.0], np.array([], dtype=np.int32)]
}
self._test(
args, expected_err_re="Field 1 in record 1 is not a valid int32: 234a")
def testOutOfRangeError(self):
args = {
"records": ["1", "9999999999999999999999999", "3"],
"record_defaults": [[1]]
}
self._test(
args, expected_err_re="Field 0 in record 1 is not a valid int32: ")
def testWrongFieldFloatError(self):
args = {
"records": [",1", "0.2,2", "3.0adf,3"],
"record_defaults": [[1.0], np.array([], dtype=np.int32)]
}
self._test(
args, expected_err_re="Field 0 in record 2 is not a valid float: ")
def testWrongFieldStringError(self):
args = {"records": ['"1,a,"', "0.22", 'a"bc'], "record_defaults": [["a"]]}
self._test(
args, expected_err_re="Unquoted fields cannot have quotes/CRLFs inside")
def testWrongDefaults(self):
args = {"records": [",1", "0.2,2", "3.0adf,3"], "record_defaults": [[1.0]]}
self._test(args, expected_err_re="Expect 1 fields but have 2 in record 0")
def testShortQuotedString(self):
args = {
"records": ["\""],
"record_defaults": [["default"]],
}
self._test(
args, expected_err_re="Quoted field has to end with quote followed.*")
def testSelectCols(self):
args = {
"records": [",,", "4,5,6"],
"record_defaults": [[1], [2]],
"select_cols": [0, 1]
}
expected_out = [[1, 4], [2, 5]]
self._test(args, expected_out)
def testSelectColsInclLast(self):
# The last col is a edge-casey; add test for that
args = {
"records": [",,", "4,5,6"],
"record_defaults": [[0], [1], [2]],
"select_cols": [0, 1, 2]
}
expected_out = [[0, 4], [1, 5], [2, 6]]
self._test(args, expected_out)
def testWrongSelectColsInclLast(self):
# The last col is a edge-casey; add test for that
args = {
"records": [",,", "4,5,6"],
"record_defaults": [[0], [1], [2]],
"select_cols": [0, 1, 3]
}
self._test(args, expected_err_re="Expect 3 fields but have 2 in record 0")
def testWrongSelectColsLen(self):
args = {
"records": ["1,2,3", "4,5,6"],
"record_defaults": [[0], [0], [0]],
"select_cols": [0]
}
with self.assertRaisesWithPredicateMatch(
ValueError, "Length of select_cols and record_defaults do not match."):
self._test(args)
def testWrongSelectColsSorting(self):
args = {
"records": ["1,2,3"],
"record_defaults": [[0], [1]],
"select_cols": [1, 0]
}
with self.assertRaisesWithPredicateMatch(
ValueError, "select_cols is not strictly increasing."):
self._test(args)
def testWrongSelectColsIndicesNegative(self):
args = {
"records": ["1,2,3"],
"record_defaults": [[0], [1]],
"select_cols": [-1, 0] # -1 is not a valid index
}
with self.assertRaisesWithPredicateMatch(
ValueError, "select_cols contains negative values."):
self._test(args)
def testWrongSelectColsIndicesTooHigh(self):
args = {
"records": ["1,2,3"],
"record_defaults": [[0], [1]],
"select_cols": [0, 3] # 3 is not a valid index
}
# Only successfully parses one of the columns
self._test(args, expected_err_re="Expect 2 fields but have 1 in record 0")
if __name__ == "__main__":
test.main()
|
