class TfTest:
_test_option = MatrixTestOptions(
file_id_content_map={
0: "ha ha ha ha la ta ha",
2: "la la ta ta da da ha",
3: "ta da ha"
},
front_end_option=MatrixFrontEndOption(
token_option=TokenOption(n_gram_size=1, token_type="word"),
norm_option=NormOption(use_freq=True,
use_tf_idf=True,
tf_idf_norm_option='l1'),
culling_option=CullingOption(cull_least_seg=None,
mfw_lowest_rank=None),
id_temp_label_map={
0: "test_label_1",
2: "test_label_2",
3: "test_label_3"
}))
model = MatrixModel(test_options=_test_option)
expected_final_dtm = pd.DataFrame(
[[0.000000, 0.675177, 0.189788, 0.135035],
[0.326024, 0.115984, 0.326024, 0.231968],
[0.412709, 0.293646, 0.000000, 0.293646]],
index=[0, 2, 3],
columns=['da', 'ha', 'la', 'ta'])
class BasicTest:
"""A class that packs all the option for a basic testing"""
_test_option = MatrixTestOptions(
file_id_content_map={
0: "ha ha ha ha la ta ha",
2: "la la ta ta da da ha",
3: "ta da ha"
},
front_end_option=MatrixFrontEndOption(
token_option=TokenOption(n_gram_size=1, token_type="word"),
norm_option=NormOption(use_freq=True,
use_tf_idf=False,
tf_idf_norm_option='l1'),
culling_option=CullingOption(cull_least_seg=None,
mfw_lowest_rank=None),
id_temp_label_map={
0: "test_label_1",
2: "test_label_2",
3: "test_label_3"
}))
model = MatrixModel(test_options=_test_option)
expected_raw_count_matrix = pd.DataFrame(
[[0., 5., 1., 1.], [2., 1., 2., 2.], [1., 1., 0., 1.]],
index=[0, 2, 3],
columns=['da', 'ha', 'la', 'ta'])
expected_final_dtm = pd.DataFrame(
[[0., .714285714286, 0.142857142857, .142857142857],
[.285714285714, .142857142857, 0.285714285714, .285714285714],
[.333333333333, .333333333333, 0., 0.333333333333]],
index=[0, 2, 3],
columns=['da', 'ha', 'la', 'ta'])
class CullingTest:
"""A class that packs all the option for a culling test"""
_test_option = MatrixTestOptions(
file_id_content_map={
1: "ha ha ha ha la ha",
2: "la la ta ta da da ha ha ha",
3: "la da ha"
},
front_end_option=MatrixFrontEndOption(
token_option=TokenOption(n_gram_size=1, token_type="word"),
norm_option=NormOption(use_freq=True,
use_tf_idf=False,
tf_idf_norm_option='l1'),
culling_option=CullingOption(cull_least_seg=3, mfw_lowest_rank=1),
id_temp_label_map={
1: "label_1",
2: "test_2",
3: "la_la_la_3"
}))
model = MatrixModel(test_options=_test_option)
expected_raw_count_matrix = pd.DataFrame(
[[0., 5., 1., 0.], [2., 3., 2., 2.], [1., 1., 1., 0.]],
index=[1, 2, 3],
columns=['da', 'ha', 'la', 'ta'])
expected_final_dtm = pd.DataFrame([[1.], [1.], [1.]],
index=[1, 2, 3],
columns=['ha'])
def _doc_term_matrix(self) -> pd.DataFrame:
""":return: the document term matrix."""
return self._test_dtm if self._test_dtm is not None \
else MatrixModel().get_matrix()
def _id_temp_label_map(self) -> IdTempLabelMap:
""":return: a map takes an id to temp labels."""
return self._test_id_temp_label_map \
if self._test_id_temp_label_map is not None \
else MatrixModel().get_id_temp_label_map()