adding deeplab_v3_plus_presets (#2051)

* adding deeplab_v3_plus_presets

* update formatting

* fixed error

* reformatted

* fix errors

* added num_classses arg

* add input_shape arg

* update test

* update input_shape

* updated preset

* update test

* move update to from_preset to task.py

* reformatted code

* updated doc string

* Update assert

* Update deeplab_v3_plus_test.py test

* Update task.py

* Update deeplab_v3_plus_test.py

* Update deeplab_v3_plus_test.py

keras_cv/models/segmentation/deeplab_v3_plus/deeplab_v3_plus.py

@@ -21,6 +21,9 @@
 from keras_cv.models.backbones.backbone_presets import (
  backbone_presets_with_weights,
 )
+from keras_cv.models.segmentation.deeplab_v3_plus.deeplab_v3_plus_presets import ( # noqa: E501
+ deeplab_v3_plus_presets,
+)
 from keras_cv.models.task import Task
 from keras_cv.utils.python_utils import classproperty
 from keras_cv.utils.train import get_feature_extractor
@@ -229,13 +232,15 @@ def from_config(cls, config):
  @classproperty
  def presets(cls):
  """Dictionary of preset names and configurations."""
- return copy.deepcopy(backbone_presets)
+ return copy.deepcopy({**backbone_presets, **deeplab_v3_plus_presets})
 
  @classproperty
  def presets_with_weights(cls):
  """Dictionary of preset names and configurations that include
  weights."""
- return copy.deepcopy(backbone_presets_with_weights)
+ return copy.deepcopy(
+ {**backbone_presets_with_weights, **deeplab_v3_plus_presets}
+ )
 
  @classproperty
  def backbone_presets(cls):

keras_cv/models/segmentation/deeplab_v3_plus/deeplab_v3_plus_presets.py

@@ -0,0 +1,43 @@
+# Copyright 2023 The KerasCV Authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""DeepLabV3Plus presets."""
+
+from keras_cv.models.backbones.resnet_v2 import resnet_v2_backbone_presets
+
+deeplab_v3_plus_presets = {
+ "deeplab_v3_plus_resnet50_pascalvoc": {
+ "metadata": {
+ "description": (
+ "DeeplabV3Plus with a ResNet50 v2 backbone. "
+ "Trained on PascalVOC 2012 Semantic segmentation task, which "
+ "consists of 20 classes and one background class. This model "
+ "achieves a final categorical accuracy of 89.34% and mIoU of "
+ "0.6391 on evaluation dataset."
+ ),
+ "params": 39191488,
+ "official_name": "DeepLabV3Plus",
+ "path": "deeplab_v3_plus",
+ },
+ "config": {
+ "backbone": resnet_v2_backbone_presets.backbone_presets[
+ "resnet50_v2_imagenet"
+ ],
+ # 21 used as an implicit background class marginally improves
+ # performance.
+ "num_classes": 21,
+ },
+ "weights_url": "https://storage.googleapis.com/keras-cv/models/deeplab_v3_plus/voc/deeplabv3plus_resenet50_pascal_voc.weights.h5", # noqa: E501
+ "weights_hash": "9681410a57bea2bc5cb7d79a1802d872ac263faab749cfe5ffdae6d6c3082041", # noqa: E501
+ },
+}

keras_cv/models/segmentation/deeplab_v3_plus/deeplab_v3_plus_test.py

@@ -74,6 +74,18 @@ def test_weights_change(self):
  self.assertNotAllEqual(w1, w2)
  self.assertFalse(ops.any(ops.isnan(w2)))
 
+ @pytest.mark.large
+ def test_with_model_preset_forward_pass(self):
+ model = DeepLabV3Plus.from_preset(
+ "deeplab_v3_plus_resnet50_pascalvoc",
+ num_classes=21,
+ input_shape=[512, 512, 3],
+ )
+ image = np.ones((1, 512, 512, 3))
+ output = ops.expand_dims(ops.argmax(model(image), axis=-1), axis=-1)
+ expected_output = np.zeros((1, 512, 512, 1))
+ self.assertAllClose(output, expected_output)
+
  @parameterized.named_parameters(
  ("tf_format", "tf", "model"),
  ("keras_format", "keras_v3", "model.keras"),

keras_cv/models/task.py

@@ -82,6 +82,7 @@ def from_preset(
  cls,
  preset,
  load_weights=None,
+ input_shape=None,
  **kwargs,
  ):
  """Instantiate {{model_name}} model from preset config and weights.
@@ -93,6 +94,9 @@ def from_preset(
  load_weights: Whether to load pre-trained weights into model.
  Defaults to `None`, which follows whether the preset has
  pretrained weights available.
+ input_shape : input shape that will be passed to backbone
+ initialization, Defaults to `None`.If `None`, the preset
+ value will be used.
 
  Examples:
  ```python
@@ -138,6 +142,8 @@ def from_preset(
 
  # Otherwise must be one of class presets
  config = metadata["config"]
+ if input_shape is not None:
+ config["backbone"]["config"]["input_shape"] = input_shape
  model = cls.from_config({**config, **kwargs})
 
  if preset not in cls.presets_with_weights or load_weights is False: