cassdlcm/mediapipe
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Internal change

Browse Source 
PiperOrigin-RevId: 538313290
This commit is contained in:
MediaPipe Team 2023-06-06 15:51:00 -07:00 committed by Copybara-Service
parent d6f34f6aef
commit 4b0f3cacae
1 changed files with 8 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
mediapipe/model_maker/python/vision/object_detector/model.py
View File

@ -101,11 +101,14 @@ class ObjectDetectorModel(tf.keras.Model):
) )
return model_config return model_config
def _build_model(self) -> tf.keras.Model: def _build_model(self, omit_l2=False) -> tf.keras.Model:
"""Builds a RetinaNet object detector model.""" """Builds a RetinaNet object detector model."""
input_specs = tf.keras.layers.InputSpec( input_specs = tf.keras.layers.InputSpec(
shape=[None] + self._model_spec.input_image_shape shape=[None] + self._model_spec.input_image_shape
) )
if omit_l2:
l2_regularizer = None
else:
l2_regularizer = tf.keras.regularizers.l2( l2_regularizer = tf.keras.regularizers.l2(
self._model_options.l2_weight_decay / 2.0 self._model_options.l2_weight_decay / 2.0
) )
@ -167,7 +170,7 @@ class ObjectDetectorModel(tf.keras.Model):
def convert_to_qat(self) -> None: def convert_to_qat(self) -> None:
"""Converts the model to a QAT RetinaNet model.""" """Converts the model to a QAT RetinaNet model."""
model = self._build_model() model = self._build_model(omit_l2=True)
dummy_input = tf.zeros([1] + self._model_spec.input_image_shape) dummy_input = tf.zeros([1] + self._model_spec.input_image_shape)
model(dummy_input, training=True) model(dummy_input, training=True)
model.set_weights(self._model.get_weights()) model.set_weights(self._model.get_weights())