tensorflow_estimator/python/estimator/canned/linear_testing_utils.py [362:466]: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - self._linear_regressor_fn = linear_regressor_fn self._fc_lib = fc_lib def setUp(self): self._model_dir = tempfile.mkdtemp() def tearDown(self): if self._model_dir: tf.compat.v1.summary.FileWriterCache.clear() shutil.rmtree(self._model_dir) def test_1d(self): """Tests predict when all variables are one-dimensional.""" with tf.Graph().as_default(): tf.Variable([[10.]], name='linear/linear_model/x/weights') tf.Variable([.2], name=BIAS_NAME) tf.Variable(100, name='global_step', dtype=tf.dtypes.int64) save_variables_to_ckpt(self._model_dir) linear_regressor = self._linear_regressor_fn( feature_columns=(self._fc_lib.numeric_column('x'),), model_dir=self._model_dir) predict_input_fn = numpy_io.numpy_input_fn( x={'x': np.array([[2.]])}, y=None, batch_size=1, num_epochs=1, shuffle=False) predictions = linear_regressor.predict(input_fn=predict_input_fn) predicted_scores = list([x['predictions'] for x in predictions]) # x * weight + bias = 2. * 10. + .2 = 20.2 self.assertAllClose([[20.2]], predicted_scores) def testMultiDim(self): """Tests predict when all variables are multi-dimenstional.""" batch_size = 2 label_dimension = 3 x_dim = 4 feature_columns = (self._fc_lib.numeric_column('x', shape=(x_dim,)),) with tf.Graph().as_default(): tf.Variable( # shape=[x_dim, label_dimension] [[1., 2., 3.], [2., 3., 4.], [3., 4., 5.], [4., 5., 6.]], name='linear/linear_model/x/weights') tf.Variable( # shape=[label_dimension] [.2, .4, .6], name=BIAS_NAME) tf.Variable(100, name='global_step', dtype=tf.dtypes.int64) save_variables_to_ckpt(self._model_dir) linear_regressor = self._linear_regressor_fn( feature_columns=feature_columns, label_dimension=label_dimension, model_dir=self._model_dir) predict_input_fn = numpy_io.numpy_input_fn( # x shape=[batch_size, x_dim] x={'x': np.array([[1., 2., 3., 4.], [5., 6., 7., 8.]])}, y=None, batch_size=batch_size, num_epochs=1, shuffle=False) predictions = linear_regressor.predict(input_fn=predict_input_fn) predicted_scores = list([x['predictions'] for x in predictions]) # score = x * weight + bias, shape=[batch_size, label_dimension] self.assertAllClose([[30.2, 40.4, 50.6], [70.2, 96.4, 122.6]], predicted_scores) def testTwoFeatureColumns(self): """Tests predict with two feature columns.""" with tf.Graph().as_default(): tf.Variable([[10.]], name='linear/linear_model/x0/weights') tf.Variable([[20.]], name='linear/linear_model/x1/weights') tf.Variable([.2], name=BIAS_NAME) tf.Variable(100, name='global_step', dtype=tf.dtypes.int64) save_variables_to_ckpt(self._model_dir) linear_regressor = self._linear_regressor_fn( feature_columns=(self._fc_lib.numeric_column('x0'), self._fc_lib.numeric_column('x1')), model_dir=self._model_dir) predict_input_fn = numpy_io.numpy_input_fn( x={ 'x0': np.array([[2.]]), 'x1': np.array([[3.]]) }, y=None, batch_size=1, num_epochs=1, shuffle=False) predictions = linear_regressor.predict(input_fn=predict_input_fn) predicted_scores = list([x['predictions'] for x in predictions]) # x0 * weight0 + x1 * weight1 + bias = 2. * 10. + 3. * 20 + .2 = 80.2 self.assertAllClose([[80.2]], predicted_scores) def testTwoFeatureColumnsMix(self): """Tests predict with two feature columns.""" with tf.Graph().as_default(): tf.Variable([[10.]], name='linear/linear_model/x0/weights') tf.Variable([[20.]], name='linear/linear_model/x1/weights') tf.Variable([.2], name=BIAS_NAME) tf.Variable(100, name='global_step', dtype=tf.dtypes.int64) save_variables_to_ckpt(self._model_dir) linear_regressor = self._linear_regressor_fn( - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - tensorflow_estimator/python/estimator/canned/v1/linear_testing_utils_v1.py [439:543]: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - self._linear_regressor_fn = linear_regressor_fn self._fc_lib = fc_lib def setUp(self): self._model_dir = tempfile.mkdtemp() def tearDown(self): if self._model_dir: tf.compat.v1.summary.FileWriterCache.clear() shutil.rmtree(self._model_dir) def test_1d(self): """Tests predict when all variables are one-dimensional.""" with tf.Graph().as_default(): tf.Variable([[10.]], name='linear/linear_model/x/weights') tf.Variable([.2], name=BIAS_NAME) tf.Variable(100, name='global_step', dtype=tf.dtypes.int64) save_variables_to_ckpt(self._model_dir) linear_regressor = self._linear_regressor_fn( feature_columns=(self._fc_lib.numeric_column('x'),), model_dir=self._model_dir) predict_input_fn = numpy_io.numpy_input_fn( x={'x': np.array([[2.]])}, y=None, batch_size=1, num_epochs=1, shuffle=False) predictions = linear_regressor.predict(input_fn=predict_input_fn) predicted_scores = list([x['predictions'] for x in predictions]) # x * weight + bias = 2. * 10. + .2 = 20.2 self.assertAllClose([[20.2]], predicted_scores) def testMultiDim(self): """Tests predict when all variables are multi-dimenstional.""" batch_size = 2 label_dimension = 3 x_dim = 4 feature_columns = (self._fc_lib.numeric_column('x', shape=(x_dim,)),) with tf.Graph().as_default(): tf.Variable( # shape=[x_dim, label_dimension] [[1., 2., 3.], [2., 3., 4.], [3., 4., 5.], [4., 5., 6.]], name='linear/linear_model/x/weights') tf.Variable( # shape=[label_dimension] [.2, .4, .6], name=BIAS_NAME) tf.Variable(100, name='global_step', dtype=tf.dtypes.int64) save_variables_to_ckpt(self._model_dir) linear_regressor = self._linear_regressor_fn( feature_columns=feature_columns, label_dimension=label_dimension, model_dir=self._model_dir) predict_input_fn = numpy_io.numpy_input_fn( # x shape=[batch_size, x_dim] x={'x': np.array([[1., 2., 3., 4.], [5., 6., 7., 8.]])}, y=None, batch_size=batch_size, num_epochs=1, shuffle=False) predictions = linear_regressor.predict(input_fn=predict_input_fn) predicted_scores = list([x['predictions'] for x in predictions]) # score = x * weight + bias, shape=[batch_size, label_dimension] self.assertAllClose([[30.2, 40.4, 50.6], [70.2, 96.4, 122.6]], predicted_scores) def testTwoFeatureColumns(self): """Tests predict with two feature columns.""" with tf.Graph().as_default(): tf.Variable([[10.]], name='linear/linear_model/x0/weights') tf.Variable([[20.]], name='linear/linear_model/x1/weights') tf.Variable([.2], name=BIAS_NAME) tf.Variable(100, name='global_step', dtype=tf.dtypes.int64) save_variables_to_ckpt(self._model_dir) linear_regressor = self._linear_regressor_fn( feature_columns=(self._fc_lib.numeric_column('x0'), self._fc_lib.numeric_column('x1')), model_dir=self._model_dir) predict_input_fn = numpy_io.numpy_input_fn( x={ 'x0': np.array([[2.]]), 'x1': np.array([[3.]]) }, y=None, batch_size=1, num_epochs=1, shuffle=False) predictions = linear_regressor.predict(input_fn=predict_input_fn) predicted_scores = list([x['predictions'] for x in predictions]) # x0 * weight0 + x1 * weight1 + bias = 2. * 10. + 3. * 20 + .2 = 80.2 self.assertAllClose([[80.2]], predicted_scores) def testTwoFeatureColumnsMix(self): """Tests predict with two feature columns.""" with tf.Graph().as_default(): tf.Variable([[10.]], name='linear/linear_model/x0/weights') tf.Variable([[20.]], name='linear/linear_model/x1/weights') tf.Variable([.2], name=BIAS_NAME) tf.Variable(100, name='global_step', dtype=tf.dtypes.int64) save_variables_to_ckpt(self._model_dir) linear_regressor = self._linear_regressor_fn( - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -