def util_batch_unbatch_array(model: Model[Floats2d, Array2d],
in_data: Floats2d, out_data: Array2d):
unbatched = [model.ops.reshape2f(a, 1, -1) for a in in_data]
with data_validation(True):
model.initialize(in_data, out_data)
Y_batched = model.predict(in_data).tolist()
Y_not_batched = [model.predict(u)[0].tolist() for u in unbatched]
assert_almost_equal(Y_batched, Y_not_batched, decimal=4)
def util_batch_unbatch_ragged(model: Model[Ragged, Array2d], in_data: Ragged,
out_data: Array2d):
with data_validation(True):
model.initialize(in_data, out_data)
Y_batched = model.predict(in_data)
Y_not_batched = [
model.predict(in_data[i])[0] for i in range(len(in_data))
]
assert_almost_equal(Y_batched, Y_not_batched, decimal=4)
def util_batch_unbatch_list(
model: Model[List[Array2d], List[Array2d]],
in_data: List[Array2d],
out_data: List[Array2d],
):
with data_validation(True):
model.initialize(in_data, out_data)
Y_batched = model.predict(in_data)
Y_not_batched = [model.predict([u])[0] for u in in_data]
assert_almost_equal(Y_batched, Y_not_batched, decimal=4)
def test_layers_from_config(name, kwargs, in_data, out_data):
cfg = {"@layers": name, **kwargs}
model = registry.resolve({"config": cfg})["config"]
if "LSTM" in name:
model = with_padded(model)
valid = True
with data_validation(valid):
model.initialize(in_data, out_data)
Y, backprop = model(in_data, is_train=True)
assert_data_match(Y, out_data)
dX = backprop(Y)
assert_data_match(dX, in_data)
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_layers_from_config(name, kwargs, in_data, out_data):
cfg = {"@layers": name, **kwargs}
filled = registry.fill_config({"config": cfg})
model = registry.make_from_config(filled)["config"]
if "LSTM" in name:
model = with_padded(model)
valid = True
if "FeatureExtractor" in name: # can't validate fake docs:
valid = False
with data_validation(valid):
model.initialize(in_data, out_data)
Y, backprop = model(in_data, is_train=True)
assert_data_match(Y, out_data)
dX = backprop(Y)
assert_data_match(dX, in_data)
def test_layers_from_config(name, kwargs, in_data, out_data):
cfg = {"@layers": name, **kwargs}
model = registry.resolve({"config": cfg})["config"]
if "LSTM" in name:
model = with_padded(model)
valid = True
with data_validation(valid):
model.initialize(in_data, out_data)
Y, backprop = model(in_data, is_train=True)
assert_data_match(Y, out_data)
dX = backprop(Y)
assert_data_match(dX, in_data)
# Test that during predictions, no dropout is applied
model._to_ops(NoDropoutOps())
model.predict(in_data)
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)
