Fix all linting errors

Author: LukeWood

examples/layers/preprocessing/cut_mix_demo.py

@@ -1,7 +1,7 @@
 """cut_mix_demo.py shows how to use the CutMix preprocessing layer.
 
-Operates on the oxford_flowers102 dataset.  In this script the flowers 
-are loaded, then are passed through the preprocessing layers.  
+Operates on the oxford_flowers102 dataset.  In this script the flowers
+are loaded, then are passed through the preprocessing layers.
 Finally, they are shown using matplotlib.
 """
 import matplotlib.pyplot as plt

examples/layers/preprocessing/mix_up_demo.py

@@ -1,7 +1,7 @@
 """mix_up_demo.py shows how to use the MixUp preprocessing layer.
 
-Uses the oxford_flowers102 dataset.  In this script the flowers 
-are loaded, then are passed through the preprocessing layers.  
+Uses the oxford_flowers102 dataset.  In this script the flowers
+are loaded, then are passed through the preprocessing layers.
 Finally, they are shown using matplotlib.
 """
 import matplotlib.pyplot as plt

keras_cv/layers/preprocessing/cut_mix.py

@@ -12,7 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import tensorflow as tf
-import tensorflow.keras as keras
 import tensorflow.keras.layers as layers
 from tensorflow.python.platform import tf_logging as logging
 
@@ -22,13 +21,14 @@ class CutMix(layers.Layer):
 
     Args:
         rate: Float between 0 and 1.  The fraction of samples to augment.
-        alpha: Float between 0 and 1.  Inverse scale parameter for the gamma distribution.
-            This controls the shape of the distribution from which the smoothing values are
-            sampled.  Defaults 1.0, which is a recommended value when training an imagenet1k
-            classification model.
-        label_smoothing: Float in [0, 1]. When > 0, label values are smoothed, meaning the
-            confidence on label values are relaxed. e.g. label_smoothing=0.2 means that we
-            will use a value of 0.1 for label 0 and 0.9 for label 1.  Defaults 0.0.
+        alpha: Float between 0 and 1.  Inverse scale parameter for the gamma
+            distribution.  This controls the shape of the distribution from which the
+            smoothing values are sampled.  Defaults 1.0, which is a recommended value
+            when training an imagenet1k classification model.
+        label_smoothing: Float in [0, 1]. When > 0, label values are smoothed,
+            meaning the confidence on label values are relaxed. e.g.
+            label_smoothing=0.2 means that we will use a value of 0.1 for label 0 and
+            0.9 for label 1.  Defaults 0.0.
     References:
        [CutMix paper]( https://arxiv.org/abs/1905.04899).
 
@@ -57,17 +57,21 @@ def call(self, images, labels):
         """call method for the CutMix layer.
 
         Args:
-            images: Tensor representing images of shape [batch_size, width, height, channels], with dtype tf.float32.
-            labels: One hot encoded tensor of labels for the images, with dtype tf.float32.
+            images: Tensor representing images of shape:
+                [batch_size, width, height, channels], with dtype tf.float32.
+            labels: One hot encoded tensor of labels for the images, with dtype
+                tf.float32.
         Returns:
             images: augmented images, same shape as input.
-            labels: updated labels with both label smoothing and the cutmix updates applied.
+            labels: updated labels with both label smoothing and the cutmix updates
+                applied.
         """
 
         if tf.shape(images)[0] == 1:
             logging.warning(
-                "CutMix received a single image to `call`.  The layer relies on combining multiple examples, "
-                "and as such will not behave as expected.  Please call the layer with 2 or more samples."
+                "CutMix received a single image to `call`.  The layer relies on "
+                "combining multiple examples, and as such will not behave as "
+                "expected.  Please call the layer with 2 or more samples."
             )
 
         augment_cond = tf.less(

keras_cv/layers/preprocessing/mix_up.py

@@ -12,7 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import tensorflow as tf
-import tensorflow.keras as keras
 import tensorflow.keras.layers as layers
 from tensorflow.python.platform import tf_logging as logging
 
@@ -22,13 +21,14 @@ class MixUp(layers.Layer):
 
     Args:
         rate: Float between 0 and 1.  The fraction of samples to augment.
-        alpha: Float between 0 and 1.  Inverse scale parameter for the gamma distribution.
-            This controls the shape of the distribution from which the smoothing values are
-            sampled.  Defaults 0.2, which is a recommended value when training an imagenet1k
-            classification model.
-        label_smoothing: Float in [0, 1]. When > 0, label values are smoothed, meaning the
-            confidence on label values are relaxed. e.g. label_smoothing=0.2 means that we
-            will use a value of 0.1 for label 0 and 0.9 for label 1.  Defaults 0.0.
+        alpha: Float between 0 and 1.  Inverse scale parameter for the gamma
+            distribution.  This controls the shape of the distribution from which the
+            smoothing values are sampled.  Defaults 0.2, which is a recommended value
+            when training an imagenet1k classification model.
+        label_smoothing: Float in [0, 1]. When > 0, label values are smoothed,
+            meaning the confidence on label values are relaxed. e.g.
+            label_smoothing=0.2 means that we will use a value of 0.1 for label 0 and
+            0.9 for label 1.  Defaults 0.0.
     References:
         [MixUp paper](https://arxiv.org/abs/1710.09412).
 
@@ -57,17 +57,21 @@ def call(self, images, labels):
         """call method for the MixUp layer.
 
         Args:
-            images: Tensor representing images of shape [batch_size, width, height, channels], with dtype tf.float32.
-            labels: One hot encoded tensor of labels for the images, with dtype tf.float32.
+            images: Tensor representing images of shape
+                [batch_size, width, height, channels], with dtype tf.float32.
+            labels: One hot encoded tensor of labels for the images, with dtype
+                tf.float32.
         Returns:
             images: augmented images, same shape as input.
-            labels: updated labels with both label smoothing and the cutmix updates applied.
+            labels: updated labels with both label smoothing and the cutmix updates
+                applied.
         """
 
         if tf.shape(images)[0] == 1:
             logging.warning(
-                "MixUp received a single image to `call`.  The layer relies on combining multiple examples, "
-                "and as such will not behave as expected.  Please call the layer with 2 or more samples."
+                "CutMix received a single image to `call`.  The layer relies on "
+                "combining multiple examples, and as such will not behave as "
+                "expected.  Please call the layer with 2 or more samples."
             )
 
         augment_cond = tf.less(

keras_cv/metrics/coco/base.py

@@ -1,3 +1,16 @@
+# Copyright 2022 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.
 import tensorflow as tf
 import tensorflow.keras as keras
 import tensorflow.keras.initializers as initializers
@@ -11,22 +24,23 @@ class COCOBase(keras.metrics.Metric):
     """COCOBase serves as a base for COCORecall and COCOPrecision.
 
     Args:
-        iou_thresholds: defaults to [0.5:0.05:0.95].  Dimension T=len(iou_thresholds), defaults to 10.
-        category_ids: no default, users must provide.  K=len(category_ids)
+        iou_thresholds: defaults to [0.5:0.05:0.95].
+        category_ids: no default, users must provide.
         area_range: area range to consider bounding boxes in. Defaults to all.
         max_detections: number of maximum detections a model is allowed to make.
 
     Internally the COCOBase class tracks the following values:
-
-    - TruePositives: tf.Tensor with shape [TxK] precision for every evaluation setting.
-    - FalsePositives: tf.Tensor with shape [TxK] precision for every evaluation setting.
-    - GroundTruthBoxes: tf.Tensor with shape [K] max recall for every evaluation setting.
+    T=len(iou_thresholds)
+    K=len(category_ids)
+    - TruePositives: tf.Tensor with shape [TxK].
+    - FalsePositives: tf.Tensor with shape [TxK].
+    - GroundTruthBoxes: tf.Tensor with shape [K].
     """
 
     def __init__(
         self,
+        category_ids,
         iou_thresholds=None,
-        category_ids=None,
         area_range=(0, 1e9 ** 2),
         max_detections=100,
         **kwargs
@@ -39,7 +53,7 @@ def __init__(
         self.iou_thresholds = self._add_constant_weight(
             "iou_thresholds", self._user_iou_thresholds
         )
-        # TODO(lukewood): support inference of category_ids based on update_state calls.
+        # TODO(lukewood): support inference of category_ids based on update_state.
         self.category_ids = self._add_constant_weight("category_ids", category_ids)
 
         self.area_range = area_range
@@ -201,8 +215,9 @@ def _match_boxes(self, y_true, y_pred, threshold, ious):
                 # TODO(lukewood): update clause to account for gtIg
                 # if m > -1 and gtIg[m] == 0 and gtIg[gind] == 1:
 
-                if not ious[gt_idx, detection_idx] >= threshold:
+                if not ious[gt_idx, detection_idx] >= iou:
                     continue
+
                 iou = ious[gt_idx, detection_idx]
                 match_index = gt_idx
 

keras_cv/metrics/coco/base_test.py

@@ -1,6 +1,18 @@
+# Copyright 2022 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.
 """Tests for bbox functions."""
 
-import numpy as np
 import tensorflow as tf
 
 from keras_cv.metrics.coco.base import COCOBase

keras_cv/metrics/coco/iou.py

@@ -1,57 +1,29 @@
+# Copyright 2022 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.
 """Contains functions to compute ious of bounding boxes."""
 import tensorflow as tf
 
-from keras_cv.util import bbox
-
-
-def _compute_single_iou(bboxes1, bb2):
-    """computes the intersection over union between bboxes1 and bb2.
-
-    bboxes1 and bb2 are expected to come in the 'corners' format, or:
-      [left, right, top, bottom].
-    For example, the bounding box with it's left side at 100, right side at 200,
-    top at 101, and bottom at 201 would be represented as:
-      [100, 200, 101, 201]
-
-    Args:
-      bboxes1: Tensor representing the first bounding box in 'corners' format.
-      bb2: Tensor representing the second bounding box in 'corners' format.
-
-    Returns:
-      iou: Tensor representing the intersection over union of the two bounding
-        boxes.
-    """
-    # bounding box indices
-
-    intersection_left = tf.math.maximum(bboxes1[bbox.LEFT], bb2[bbox.LEFT])
-    intersection_right = tf.math.minimum(bboxes1[bbox.RIGHT], bb2[bbox.RIGHT])
-
-    intersection_top = tf.math.maximum(bboxes1[bbox.TOP], bb2[bbox.TOP])
-    intersection_bottom = tf.math.minimum(bboxes1[bbox.BOTTOM], bb2[bbox.BOTTOM])
-
-    area_intersection = _area(
-        intersection_left, intersection_right, intersection_top, intersection_bottom
-    )
-    area_bboxes1 = _area(
-        bb1[bbox.LEFT], bb1[bbox.RIGHT], bb1[bbox.TOP], bb1[bbox.BOTTOM]
-    )
-    area_bb2 = _area(
-        bb2[bbox.LEFT], bb2[bbox.RIGHT], bb2[bbox.TOP], bb2[bbox.BOTTOM]
-    )
-
-    area_union = area_bboxes1 + area_bb2 - area_intersection
-    return tf.math.divide_no_nan(area_intersection, area_union)
-
 
 def compute_ious_for_image(boxes1, boxes2):
     """computes a lookup table vector containing the ious for a given set boxes.
 
     The lookup vector is to be indexed by [`boxes1_index`,`boxes2_index`].
 
-    Bounding boxes are expected to be in the corners format of `[bbox.LEFT, bbox.RIGHT,
-    bbox.TOP, bbox.BOTTOM]`.  For example, the bounding box with it's left side at 100,
-    bbox.RIGHT side at 200, bbox.TOP at 101, and bbox.BOTTOM at 201 would be represented
-    as:
+    Bounding boxes are expected to be in the corners format of
+    `[bbox.LEFT, bbox.RIGHT, bbox.TOP, bbox.BOTTOM]`.  For example, the bounding box
+    with it's left side at 100, bbox.RIGHT side at 200, bbox.TOP at 101, and
+    bbox.BOTTOM at 201 would be represented as:
     > [100, 200, 101, 201]
 
     Args:

keras_cv/metrics/coco/iou_test.py

@@ -1,3 +1,16 @@
+# Copyright 2022 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.
 """Tests for iou functions."""
 
 import numpy as np

keras_cv/metrics/coco/numerical_tests/recall_correctness_test.py

@@ -18,28 +18,28 @@
 import numpy as np
 import tensorflow as tf
 
-from keras_cv.metrics.coco import iou as iou_lib
 from keras_cv.metrics.coco.recall import COCORecall
 from keras_cv.util import bbox
 
 SAMPLE_FILE = os.path.dirname(os.path.abspath(__file__)) + "/sample_boxes.npz"
 
 
 class RecallCorrectnessTest(tf.test.TestCase):
-    """Unit tests that test Keras COCO metric results against the known good ones of cocoeval.py.
-    The bounding boxes in sample_boxes.npz were given to cocoeval.py which output the following values:
-         Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.643
-         Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 1.000
-         Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.729
-         Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.644
-         Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.633
-         Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.689
-         Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.504
-         Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.660
-         Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.660
-         Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.652
-         Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.644
-         Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.696
+    """Unit tests that test Keras COCO metric results against the known good ones of
+    cocoeval.py.  The bounding boxes in sample_boxes.npz were given to cocoeval.py
+    which output the following values:
+    Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.643
+    Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 1.000
+    Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.729
+    Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.644
+    Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.633
+    Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.689
+    Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.504
+    Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.660
+    Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.660
+    Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.652
+    Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.644
+    Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.696
     """
 
     def test_recall_correctness_maxdets_1(self):

keras_cv/metrics/coco/recall.py

@@ -1,3 +1,16 @@
+# Copyright 2022 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.
 import tensorflow as tf
 
 from keras_cv.metrics.coco.base import COCOBase