from keras import Model
from keras.layers import Dense
from sklearn_pandas import DataFrameMapper
from keras_pandas import lib
from keras_pandas.data_types.Text import Text
from tests.testbase import TestBase
class TestNumerical(TestBase):
def test_init(self):
# Create datatype
datatype = Text()
# Check for output support (or not)
self.assertFalse(datatype.supports_output)
def test_datatype_signature(self):
# Create datatype
datatype = Text()
# Check valid datatype
lib.check_valid_datatype(datatype)
def test_whole(self):
datatype = Text()
# Load observations
observations = lib.load_titanic()
# Transform observations
mapper = DataFrameMapper([(['name'], datatype.default_transformation_pipeline),
(['fare'], None)], df_out=True)
transformed_df = mapper.fit_transform(observations)
# Create network
input_layer, input_nub = datatype.input_nub_generator('name', transformed_df)
output_nub = Dense(1)
x = input_nub
x = output_nub(x)
model = Model(input_layer, x)
model.compile(optimizer='adam', loss='mse')
pass
