forked from goncalomcorreia/pytorch-lecture
-
Notifications
You must be signed in to change notification settings - Fork 5
/
spec.py
182 lines (164 loc) · 4.79 KB
/
spec.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
import torch
from torchtyping import TensorType as TT
from hypothesis.extra.numpy import arrays
from hypothesis.strategies import integers, lists, composite, floats
from hypothesis import given
import numpy as np
import random
import sys
import typing
tensor = torch.tensor
numpy_to_torch_dtype_dict = {
bool: torch.bool,
np.uint8: torch.uint8,
np.int8: torch.int8,
np.int16: torch.int16,
np.int32: torch.int32,
np.int64: torch.int64,
np.float16: torch.float16,
np.float32: torch.float32,
np.float64: torch.float64,
}
torch_to_numpy_dtype_dict = {v: k for k, v in numpy_to_torch_dtype_dict.items()}
@composite
def spec(draw, x, min_size=1):
# Get the type hints.
if sys.version_info >= (3, 9):
gth = typing.get_type_hints(x, include_extras=True)
else:
gth = typing.get_type_hints(x)
# Collect all the dimension names.
names = set()
for k in gth:
if not hasattr(gth[k], "__metadata__"):
continue
dims = gth[k].__metadata__[0]["details"][0].dims
names.update([d.name for d in dims if isinstance(d.name, str)])
names = list(names)
# draw sizes for each dim.
size = integers(min_value=min_size, max_value=5)
arr = draw(arrays(shape=(len(names),), unique=True, elements=size, dtype=np.int32)).tolist()
sizes = dict(zip(names, arr))
for n in list(sizes.keys()):
if '*' in n or '+' in n or '-' in n or '//' in n:
i, op, j = n.split()
i_val = i if i.isdigit() else sizes[i]
j_val = j if j.isdigit() else sizes[j]
sizes[n] = eval('{}{}{}'.format(i_val, op,j_val))
# Create tensors for each size.
ret = {}
for k in gth:
if not hasattr(gth[k], "__metadata__"):
continue
shape = tuple(
[
sizes[d.name] if isinstance(d.name, str) else d.size
for d in gth[k].__metadata__[0]["details"][0].dims
]
)
dtype = (torch_to_numpy_dtype_dict[
gth[k].__metadata__[0]["details"][1].dtype
]
if len(gth[k].__metadata__[0]["details"]) >= 2
else int)
ret[k] = draw(
arrays(
shape=shape,
dtype=dtype,
elements=integers(min_value=-5, max_value=5) if
dtype == int else None,
unique=False
)
)
ret[k] = np.nan_to_num(ret[k], nan=0, neginf=0, posinf=0)
ret["return"][:] = 0
return ret, sizes
def make_test(name, problem, problem_spec, add_sizes=[], constraint=lambda d, sizes: d):
examples = []
for i in range(3):
example, sizes = spec(problem, 3).example()
example = constraint(example, sizes=sizes)
out = example["return"].tolist()
del example["return"]
problem_spec(*example.values(), out)
for size in add_sizes:
example[size] = sizes[size]
yours = None
try:
yours = problem(*map(tensor, example.values()))
except AssertionError:
pass
for size in add_sizes:
del example[size]
example["target"] = tensor(out)
if yours is not None:
example["yours"] = yours
examples.append(example)
@given(spec(problem))
def test_problem(d):
d, sizes = d
d = constraint(d, sizes=sizes)
out = d["return"].tolist()
del d["return"]
problem_spec(*d.values(), out)
for size in add_sizes:
d[size] = sizes[size]
out2 = problem(*map(tensor, d.values()))
out = tensor(out)
out2 = torch.broadcast_to(out2, out.shape)
assert torch.allclose(
out, out2
), "Two tensors are not equal\n Spec: \n\t%s \n\t%s" % (out, out2)
return test_problem
def run_test(fn):
fn()
# Generate a random puppy video if you are correct.
print("Correct!")
from IPython.display import HTML
pups = [
"2m78jPG",
"pn1e9TO",
"MQCIwzT",
"udLK6FS",
"ZNem5o3",
"DS2IZ6K",
"aydRUz8",
"MVUdQYK",
"kLvno0p",
"wScLiVz",
"Z0TII8i",
"F1SChho",
"9hRi2jN",
"lvzRF3W",
"fqHxOGI",
"1xeUYme",
"6tVqKyM",
"CCxZ6Wr",
"lMW0OPQ",
"wHVpHVG",
"Wj2PGRl",
"HlaTE8H",
"k5jALH0",
"3V37Hqr",
"Eq2uMTA",
"Vy9JShx",
"g9I2ZmK",
"Nu4RH7f",
"sWp0Dqd",
"bRKfspn",
"qawCMl5",
"2F6j2B4",
"fiJxCVA",
"pCAIlxD",
"zJx2skh",
"2Gdl1u7",
"aJJAY4c",
"ros6RLC",
"DKLBJh7",
"eyxH0Wc",
"rJEkEw4"]
return HTML("""
<video alt="test" controls autoplay=1 height="240">
<source src="https://openpuppies.com/mp4/%s.mp4" type="video/mp4">
</video>
"""%(random.sample(pups, 1)[0]))