def test_dense_block():
utils.block_basic_exam(
basic.DenseBlock(),
tf.keras.Input(shape=(32,), dtype=tf.float32),
[
'num_layers',
'use_batchnorm',
])
def test_embedding_block():
utils.block_basic_exam(
basic.Embedding(),
tf.keras.Input(shape=(32,), dtype=tf.float32),
[
'pretraining',
'embedding_dim',
])
def test_resnet_block(init, build):
utils.block_basic_exam(
basic.ResNetBlock(),
tf.keras.Input(shape=(32, 32, 3), dtype=tf.float32),
['version', 'pooling'],
)
assert init.called
assert build.called
def test_merge():
utils.block_basic_exam(
reduction.Merge(),
[
tf.keras.Input(shape=(32, ), dtype=tf.float32),
tf.keras.Input(shape=(4, 8), dtype=tf.float32),
],
['merge_type'],
)
def test_rnn_block():
utils.block_basic_exam(
basic.RNNBlock(),
tf.keras.Input(shape=(32, 10), dtype=tf.float32),
[
'bidirectional',
'layer_type',
'num_layers',
])
def test_categorical_to_numerical():
block = preprocessing.CategoricalToNumerical()
block.column_names = ['a']
block.column_types = {'a': 'num'}
utils.block_basic_exam(
block,
tf.keras.Input(shape=(1, ), dtype=tf.string),
[],
)
def test_conv_block():
utils.block_basic_exam(
basic.ConvBlock(),
tf.keras.Input(shape=(32, 32, 3), dtype=tf.float32),
[
'kernel_size',
'num_blocks',
'separable',
])
def test_xception_block():
utils.block_basic_exam(
basic.XceptionBlock(),
tf.keras.Input(shape=(32, 32, 3), dtype=tf.float32),
[
'activation',
'initial_strides',
'num_residual_blocks',
'pooling',
])
def test_image_block():
utils.block_basic_exam(
wrapper.ImageBlock(normalize=None, augment=None),
tf.keras.Input(shape=(32, 32, 3), dtype=tf.float32),
[
'block_type',
'normalize',
'augment',
],
)
def test_timeseries_block():
block = wrapper.TimeseriesBlock()
block.column_names = ['0', '1']
block.column_types = {
'0': adapters.NUMERICAL,
'1': adapters.NUMERICAL,
}
outputs = utils.block_basic_exam(
block,
tf.keras.Input(shape=(32, 2), dtype=tf.float32),
[],
)
assert isinstance(nest.flatten(outputs)[0], tf.Tensor)
def test_structured_data_block():
block = wrapper.StructuredDataBlock()
block.column_names = ['0', '1']
block.column_types = {
'0': adapters.NUMERICAL,
'1': adapters.NUMERICAL,
}
outputs = utils.block_basic_exam(
block,
tf.keras.Input(shape=(2, ), dtype=tf.string),
[],
)
assert isinstance(nest.flatten(outputs)[0], tf.Tensor)
def test_spatial_reduction():
utils.block_basic_exam(
reduction.SpatialReduction(),
tf.keras.Input(shape=(32, 32, 3), dtype=tf.float32),
['reduction_type'],
)
def test_temporal_reduction():
utils.block_basic_exam(
reduction.TemporalReduction(),
tf.keras.Input(shape=(32, 10), dtype=tf.float32),
['reduction_type'],
)
def test_text_block():
utils.block_basic_exam(
wrapper.TextBlock(),
tf.keras.Input(shape=(1, ), dtype=tf.string),
['vectorizer'],
)
def test_text_to_int_sequence():
utils.block_basic_exam(
preprocessing.TextToIntSequence(),
tf.keras.Input(shape=(1, ), dtype=tf.string),
['output_sequence_length'],
)
def test_text_to_ngram_vector():
utils.block_basic_exam(
preprocessing.TextToNgramVector(),
tf.keras.Input(shape=(1, ), dtype=tf.string),
['ngrams'],
)
def test_image_augmentation():
utils.block_basic_exam(
preprocessing.ImageAugmentation(),
tf.keras.Input(shape=(32, 32, 3), dtype=tf.float32),
['vertical_flip', 'horizontal_flip'],
)
def test_resnet_block():
utils.block_basic_exam(
basic.ResNetBlock(),
tf.keras.Input(shape=(32, 32, 3), dtype=tf.float32),
['version', 'pooling'],
)
