Revert "Weights densenet (keras-team#1855)"

This reverts commit f67b4db.

keras_hub/api/layers/__init__.py

@@ -37,9 +37,6 @@
     ResizingImageConverter,
 )
 from keras_hub.src.layers.preprocessing.start_end_packer import StartEndPacker
-from keras_hub.src.models.densenet.densenet_image_converter import (
-    DenseNetImageConverter,
-)
 from keras_hub.src.models.pali_gemma.pali_gemma_image_converter import (
     PaliGemmaImageConverter,
 )

keras_hub/api/models/__init__.py

@@ -89,9 +89,6 @@
 from keras_hub.src.models.densenet.densenet_image_classifier import (
     DenseNetImageClassifier,
 )
-from keras_hub.src.models.densenet.densenet_image_classifier_preprocessor import (
-    DenseNetImageClassifierPreprocessor,
-)
 from keras_hub.src.models.distil_bert.distil_bert_backbone import (
     DistilBertBackbone,
 )

keras_hub/src/models/densenet/densenet_backbone.py

@@ -79,22 +79,19 @@ def __init__(
                 channel_axis,
                 stackwise_num_repeats[stack_index],
                 growth_rate,
-                name=f"stack{stack_index+1}",
+                name=f"conv{index}",
             )
             pyramid_outputs[f"P{index}"] = x
             x = apply_transition_block(
-                x,
-                channel_axis,
-                compression_ratio,
-                name=f"transition{stack_index+1}",
+                x, channel_axis, compression_ratio, name=f"pool{index}"
             )
 
         x = apply_dense_block(
             x,
             channel_axis,
             stackwise_num_repeats[-1],
             growth_rate,
-            name=f"stack{len(stackwise_num_repeats)}",
+            name=f"conv{len(stackwise_num_repeats) + 1}",
         )
         pyramid_outputs[f"P{len(stackwise_num_repeats) + 1}"] = x
         x = keras.layers.BatchNormalization(
@@ -138,7 +135,7 @@ def apply_dense_block(x, channel_axis, num_repeats, growth_rate, name=None):
 
     for i in range(num_repeats):
         x = apply_conv_block(
-            x, channel_axis, growth_rate, name=f"{name}_block{i+1}"
+            x, channel_axis, growth_rate, name=f"{name}_block_{i}"
         )
     return x
 
@@ -186,9 +183,9 @@ def apply_conv_block(x, channel_axis, growth_rate, name=None):
 
     shortcut = x
     x = keras.layers.BatchNormalization(
-        axis=channel_axis, epsilon=BN_EPSILON, name=f"{name}_1_bn"
+        axis=channel_axis, epsilon=BN_EPSILON, name=f"{name}_0_bn"
     )(x)
-    x = keras.layers.Activation("relu", name=f"{name}_1_relu")(x)
+    x = keras.layers.Activation("relu", name=f"{name}_0_relu")(x)
     x = keras.layers.Conv2D(
         4 * growth_rate,
         1,
@@ -197,9 +194,9 @@ def apply_conv_block(x, channel_axis, growth_rate, name=None):
         name=f"{name}_1_conv",
     )(x)
     x = keras.layers.BatchNormalization(
-        axis=channel_axis, epsilon=BN_EPSILON, name=f"{name}_2_bn"
+        axis=channel_axis, epsilon=BN_EPSILON, name=f"{name}_1_bn"
     )(x)
-    x = keras.layers.Activation("relu", name=f"{name}_2_relu")(x)
+    x = keras.layers.Activation("relu", name=f"{name}_1_relu")(x)
     x = keras.layers.Conv2D(
         growth_rate,
         3,

keras_hub/src/models/densenet/densenet_image_classifier.py

@@ -2,9 +2,6 @@
 
 from keras_hub.src.api_export import keras_hub_export
 from keras_hub.src.models.densenet.densenet_backbone import DenseNetBackbone
-from keras_hub.src.models.densenet.densenet_image_classifier_preprocessor import (
-    DenseNetImageClassifierPreprocessor,
-)
 from keras_hub.src.models.image_classifier import ImageClassifier
 
 
@@ -22,13 +19,7 @@ class DenseNetImageClassifier(ImageClassifier):
         num_classes: int. The number of classes to predict.
         activation: `None`, str or callable. The activation function to use on
             the `Dense` layer. Set `activation=None` to return the output
-            logits. Defaults to `None`.
-        pooling: A pooling layer to use before the final classification layer,
-            must be one of "avg" or "max". Use "avg" for
-            `GlobalAveragePooling2D` and "max" for "GlobalMaxPooling2D.
-        preprocessor: A `keras_hub.models.DenseNetImageClassifierPreprocessor`
-            or `None`. If `None`, this model will not apply preprocessing, and
-            inputs should be preprocessed before calling the model.
+            logits. Defaults to `"softmax"`.
 
     Examples:
 
@@ -82,29 +73,18 @@ class DenseNetImageClassifier(ImageClassifier):
     """
 
     backbone_cls = DenseNetBackbone
-    preprocessor_cls = DenseNetImageClassifierPreprocessor
 
     def __init__(
         self,
         backbone,
         num_classes,
-        activation=None,
-        pooling="avg",
-        preprocessor=None,
+        activation="softmax",
+        preprocessor=None,  # adding this dummy arg for saved model test
+        # TODO: once preprocessor flow is figured out, this needs to be updated
         **kwargs,
     ):
         # === Layers ===
         self.backbone = backbone
-        self.preprocessor = preprocessor
-        if pooling == "avg":
-            self.pooler = keras.layers.GlobalAveragePooling2D()
-        elif pooling == "max":
-            self.pooler = keras.layers.GlobalMaxPooling2D()
-        else:
-            raise ValueError(
-                "Unknown `pooling` type. Polling should be either `'avg'` or "
-                f"`'max'`. Received: pooling={pooling}."
-            )
         self.output_dense = keras.layers.Dense(
             num_classes,
             activation=activation,
@@ -114,7 +94,6 @@ def __init__(
         # === Functional Model ===
         inputs = self.backbone.input
         x = self.backbone(inputs)
-        x = self.pooler(x)
         outputs = self.output_dense(x)
         super().__init__(
             inputs=inputs,
@@ -125,7 +104,6 @@ def __init__(
         # === Config ===
         self.num_classes = num_classes
         self.activation = activation
-        self.pooling = pooling
 
     def get_config(self):
         # Backbone serialized in `super`
@@ -134,7 +112,6 @@ def get_config(self):
             {
                 "num_classes": self.num_classes,
                 "activation": self.activation,
-                "pooling": self.pooling,
             }
         )
         return config

keras_hub/src/models/densenet/densenet_image_classifier_test.py

@@ -23,7 +23,6 @@ def setUp(self):
             "backbone": self.backbone,
             "num_classes": 2,
             "activation": "softmax",
-            "pooling": "avg",
         }
         self.train_data = (
             self.images,
@@ -41,33 +40,10 @@ def test_classifier_basics(self):
             expected_output_shape=(2, 2),
         )
 
-    @pytest.mark.large
-    def test_smallest_preset(self):
-        # Test that our forward pass is stable!
-        image_batch = self.load_test_image()[None, ...] / 255.0
-        self.run_preset_test(
-            cls=DenseNetImageClassifier,
-            preset="densenet_121_imagenet",
-            input_data=image_batch,
-            expected_output_shape=(1, 1000),
-            expected_labels=[85],
-        )
-
     @pytest.mark.large
     def test_saved_model(self):
         self.run_model_saving_test(
             cls=DenseNetImageClassifier,
             init_kwargs=self.init_kwargs,
             input_data=self.images,
         )
-
-    @pytest.mark.extra_large
-    def test_all_presets(self):
-        for preset in DenseNetImageClassifier.presets:
-            self.run_preset_test(
-                cls=DenseNetImageClassifier,
-                preset=preset,
-                init_kwargs={"num_classes": 2},
-                input_data=self.images,
-                expected_output_shape=(2, 2),
-            )

keras_hub/src/utils/timm/preset_loader.py

@@ -3,7 +3,6 @@
 from keras_hub.src.models.image_classifier import ImageClassifier
 from keras_hub.src.utils.preset_utils import PresetLoader
 from keras_hub.src.utils.preset_utils import jax_memory_cleanup
-from keras_hub.src.utils.timm import convert_densenet
 from keras_hub.src.utils.timm import convert_resnet
 from keras_hub.src.utils.transformers.safetensor_utils import SafetensorLoader
 
@@ -14,8 +13,6 @@ def __init__(self, preset, config):
         architecture = self.config["architecture"]
         if "resnet" in architecture:
             self.converter = convert_resnet
-        if "densenet" in architecture:
-            self.converter = convert_densenet
         else:
             raise ValueError(
                 "KerasHub has no converter for timm models "