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
|
import pytest
from thinc.api import chain, Relu, reduce_max, Softmax, with_ragged
from thinc.api import ParametricAttention, list2ragged, reduce_sum
from thinc.util import DataValidationError, data_validation
def test_validation():
model = chain(Relu(10), Relu(10), with_ragged(reduce_max()), Softmax())
with data_validation(True):
with pytest.raises(DataValidationError):
model.initialize(X=model.ops.alloc2f(1, 10), Y=model.ops.alloc2f(1, 10))
with pytest.raises(DataValidationError):
model.initialize(X=model.ops.alloc3f(1, 10, 1), Y=model.ops.alloc2f(1, 10))
with pytest.raises(DataValidationError):
model.initialize(X=[model.ops.alloc2f(1, 10)], Y=model.ops.alloc2f(1, 10))
def test_validation_complex():
good_model = chain(list2ragged(), reduce_sum(), Relu(12, dropout=0.5), Relu(1))
X = [good_model.ops.xp.zeros((4, 75), dtype="f")]
Y = good_model.ops.xp.zeros((1,), dtype="f")
good_model.initialize(X, Y)
good_model.predict(X)
bad_model = chain(
list2ragged(),
reduce_sum(),
Relu(12, dropout=0.5),
# ERROR: Why can't I attach a Relu to an attention layer?
ParametricAttention(12),
Relu(1),
)
with data_validation(True):
with pytest.raises(DataValidationError):
bad_model.initialize(X, Y)
|
