test(tailor): add test for name is none

finetuner/tailor/pytorch/__init__.py

@@ -105,6 +105,7 @@ def hook(module, input, output):
                 not output_shape
                 or len(output_shape) != 2
                 or not is_list_int(output_shape)
+                or summary[layer]['cls_name'] in self._model.__class__.__name__
             ):
                 continue
 

tests/unit/tailor/test_keras.py

@@ -113,6 +113,11 @@ def test_trim_fail_given_unexpected_layer_name(model, layer_name):
         ('vgg16_cnn_model', 'fc1', (None, 4096)),
         ('stacked_lstm', 'lstm_2', (None, 256)),
         ('bidirectional_lstm', 'bidirectional', (None, 128)),
+        ('dense_model', None, (None, 10)),
+        ('simple_cnn_model', None, (None, 10)),
+        ('vgg16_cnn_model', None, (None, 1000)),
+        ('stacked_lstm', None, (None, 5)),
+        ('bidirectional_lstm', None, (None, 32)),
     ],
     indirect=['model'],
 )
@@ -159,7 +164,6 @@ def test_keras_model_parser():
     assert r[1]['name'] == 'l2'
     assert r[2]['name'] == 'l3'
 
-    # flat layer can be a nonparametric candidate
     assert r[0]['output_features'] == 784
     assert r[0]['params'] == 0
 

tests/unit/tailor/test_paddle.py

@@ -200,6 +200,25 @@ def test_freeze(model, layer_name, input_size, input_dtype):
         ),
         ('stacked_lstm', 'linear_33', (128,), (1, 128), 'int64', [1, 256]),
         ('bidirectional_lstm', 'linear_35', (128,), (1, 128), 'int64', [1, 128]),
+        ('dense_model', None, (128,), (1, 128), 'float32', [1, 10]),
+        (
+            'simple_cnn_model',
+            None,
+            (1, 28, 28),
+            (1, 1, 28, 28),
+            'float32',
+            [1, 10],
+        ),
+        (
+            'vgg16_cnn_model',
+            None,
+            (3, 224, 224),
+            (1, 3, 224, 224),
+            'float32',
+            [1, 4096],
+        ),
+        ('stacked_lstm', None, (128,), (1, 128), 'int64', [1, 5]),
+        ('bidirectional_lstm', None, (128,), (1, 128), 'int64', [1, 128]),
     ],
     indirect=['model'],
 )

tests/unit/tailor/test_torch.py

@@ -164,6 +164,25 @@ def test_trim_fail_given_unexpected_layer_idx(
         ),
         ('stacked_lstm', 'linear_3', (128,), (1, 128), 'int64', [1, 256]),
         ('bidirectional_lstm', 'linear_4', (128,), (1, 128), 'int64', [1, 128]),
+        ('dense_model', None, (128,), (1, 128), 'float32', [1, 10]),
+        (
+            'simple_cnn_model',
+            None,
+            (1, 28, 28),
+            (1, 1, 28, 28),
+            'float32',
+            [1, 10],
+        ),
+        (
+            'vgg16_cnn_model',
+            None,
+            (3, 224, 224),
+            (1, 3, 224, 224),
+            'float32',
+            [1, 4096],
+        ),
+        ('stacked_lstm', None, (128,), (1, 128), 'int64', [1, 5]),
+        ('bidirectional_lstm', None, (128,), (1, 128), 'int64', [1, 128]),
     ],
     indirect=['model'],
 )
@@ -182,7 +201,7 @@ def test_trim(
     if input_dtype == 'int64':
         input_ = input_.type(torch.IntTensor)
     out = pytorch_tailor.model(input_)
-    assert list(out.size()) == expected_output_shape  # 4th layer Linear
+    assert list(out.size()) == expected_output_shape
 
 
 @pytest.mark.parametrize(