Release MobileDet code and model, and require tf_slim installation for OD API. (#8562)

* Merged commit includes the following changes:
311933687  by Sergio Guadarrama:

    Removes spurios use of tf.compat.v2, which results in spurious tf.compat.v1.compat.v2. Adds basic test to nasnet_utils.
    Replaces all remaining import tensorflow as tf with import tensorflow.compat.v1 as tf

--
311766063  by Sergio Guadarrama:

    Removes explicit tf.compat.v1 in all call sites (we already import tf.compat.v1, so this code was  doing tf.compat.v1.compat.v1). The existing code worked in latest version of tensorflow, 2.2, (and 1.15) but not in 1.14 or in 2.0.0a, this CL fixes it.

--
311624958  by Sergio Guadarrama:

    Updates README that doesn't render properly in github documentation

--
310980959  by Sergio Guadarrama:

    Moves research_models/slim off tf.contrib.slim/layers/framework to tf_slim

--
310263156  by Sergio Guadarrama:

    Adds model breakdown for MobilenetV3

--
308640516  by Sergio Guadarrama:

    Internal change

308244396  by Sergio Guadarrama:

    GroupNormalization support for MobilenetV3.

--
307475800  by Sergio Guadarrama:

    Internal change

--
302077708  by Sergio Guadarrama:

    Remove `disable_tf2` behavior from slim py_library targets

--
301208453  by Sergio Guadarrama:

    Automated refactoring to make code Python 3 compatible.

--
300816672  by Sergio Guadarrama:

    Internal change

299433840  by Sergio Guadarrama:

    Internal change

299221609  by Sergio Guadarrama:

    Explicitly disable Tensorflow v2 behaviors for all TF1.x binaries and tests

--
299179617  by Sergio Guadarrama:

    Internal change

299040784  by Sergio Guadarrama:

    Internal change

299036699  by Sergio Guadarrama:

    Internal change

298736510  by Sergio Guadarrama:

    Internal change

298732599  by Sergio Guadarrama:

    Internal change

298729507  by Sergio Guadarrama:

    Internal change

298253328  by Sergio Guadarrama:

    Internal change

297788346  by Sergio Guadarrama:

    Internal change

297785278  by Sergio Guadarrama:

    Internal change

297783127  by Sergio Guadarrama:

    Internal change

297725870  by Sergio Guadarrama:

    Internal change

297721811  by Sergio Guadarrama:

    Internal change

297711347  by Sergio Guadarrama:

    Internal change

297708059  by Sergio Guadarrama:

    Internal change

297701831  by Sergio Guadarrama:

    Internal change

297700038  by Sergio Guadarrama:

    Internal change

297670468  by Sergio Guadarrama:

    Internal change.

--
297350326  by Sergio Guadarrama:

    Explicitly replace "import tensorflow" with "tensorflow.compat.v1" for TF2.x migration

--
297201668  by Sergio Guadarrama:

    Explicitly replace "import tensorflow" with "tensorflow.compat.v1" for TF2.x migration

--
294483372  by Sergio Guadarrama:

    Internal change

PiperOrigin-RevId: 311933687

* Merged commit includes the following changes:
312578615  by Menglong Zhu:

    Modify the LSTM feature extractors to be python 3 compatible.

--
311264357  by Menglong Zhu:

    Removes contrib.slim

--
308957207  by Menglong Zhu:

    Automated refactoring to make code Python 3 compatible.

--
306976470  by yongzhe:

    Internal change

306777559  by Menglong Zhu:

    Internal change

--
299232507  by lzyuan:

    Internal update.

--
299221735  by lzyuan:

    Add small epsilon on max_range for quantize_op to prevent range collapse.

--

PiperOrigin-RevId: 312578615

* Merged commit includes the following changes:
310447280  by lzc:

    Internal changes.

--

PiperOrigin-RevId: 310447280
Co-authored-by: Sergio Guadarrama <sguada@google.com>
Co-authored-by: Menglong Zhu <menglong@google.com>

research/object_detection/predictors/rfcn_keras_box_predictor_test.py

@@ -15,7 +15,7 @@
 
 """Tests for object_detection.predictors.rfcn_box_predictor."""
 import numpy as np
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 from google.protobuf import text_format
 from object_detection.builders import hyperparams_builder

research/object_detection/protos/input_reader.proto

@@ -24,7 +24,14 @@ enum InstanceMaskType {
   PNG_MASKS = 2;        // Encoded PNG masks.
 }
 
-// Next id: 29
+// Input type format: whether inputs are TfExamples or TfSequenceExamples.
+enum InputType {
+  INPUT_DEFAULT = 0;          // Default implementation, currently TF_EXAMPLE
+  TF_EXAMPLE = 1;  // TfExample input
+  TF_SEQUENCE_EXAMPLE = 2;        // TfSequenceExample Input
+}
+
+// Next id: 31
 message InputReader {
   // Name of input reader. Typically used to describe the dataset that is read
   // by this input reader.
@@ -119,15 +126,22 @@ message InputReader {
   // Whether to include the source_id string in the input features.
   optional bool include_source_id = 27 [default = false];
 
+  // Whether input data type is tf.Examples or tf.SequenceExamples
+  optional InputType input_type = 30 [default = TF_EXAMPLE];
+
   oneof input_reader {
     TFRecordInputReader tf_record_input_reader = 8;
     ExternalInputReader external_input_reader = 9;
   }
 
 
+  // Expand labels to ancestors or descendants in the hierarchy for
+  // for positive and negative labels, respectively.
+  optional bool expand_labels_hierarchy = 29 [default = false];
 }
 
-// An input reader that reads TF Example protos from local TFRecord files.
+// An input reader that reads TF Example or TF Sequence Example protos from
+// local TFRecord files.
 message TFRecordInputReader {
   // Path(s) to `TFRecordFile`s.
   repeated string input_path = 1;

research/object_detection/protos/optimizer.proto

@@ -37,7 +37,7 @@ message MomentumOptimizer {
 message AdamOptimizer {
   optional LearningRate learning_rate = 1;
   // Default value for epsilon (1e-8) matches default value in
-  // tf.compat.v1.train.AdamOptimizer. This differs from tf2 default of 1e-7
+  // tf.train.AdamOptimizer. This differs from tf2 default of 1e-7
   // in tf.keras.optimizers.Adam .
   optional float epsilon = 2 [default = 1e-8];
 }
@@ -67,8 +67,8 @@ message ExponentialDecayLearningRate {
   optional float decay_factor = 3 [default = 0.95];
   optional bool staircase = 4 [default = true];
   optional float burnin_learning_rate = 5 [default = 0.0];
-  optional uint32 burnin_steps =  6 [default = 0];
-  optional float min_learning_rate =  7 [default = 0.0];
+  optional uint32 burnin_steps = 6 [default = 0];
+  optional float min_learning_rate = 7 [default = 0.0];
 }
 
 // Configuration message for a manually defined learning rate schedule.

research/object_detection/protos/post_processing.proto

@@ -49,6 +49,14 @@ message BatchNonMaxSuppression {
   // Whether to change coordinate frame of the boxlist to be relative to
   // window's frame.
   optional bool change_coordinate_frame = 12 [default = true];
+
+  // Use hard NMS. Note that even if this field is set false, the behavior of
+  // NMS will be equivalent to hard NMS; This field when set to true forces the
+  // tf.image.non_max_suppression function to be called instead
+  // of tf.image.non_max_suppression_with_scores and can be used to
+  // export models for older versions of TF.
+  optional bool use_hard_nms = 13 [default = false];
+
 }
 
 // Configuration proto for post-processing predicted boxes and

research/object_detection/protos/string_int_label_map.proto

@@ -33,6 +33,11 @@ message StringIntLabelMapItem {
     optional string label = 2;
   }
   repeated KeypointMap keypoints = 4;
+
+  // Label ids for the elements that are connected in the hierarchy with the
+  // current element. Value should correspond to another label id element.
+  repeated int32 ancestor_ids = 5;
+  repeated int32 descendant_ids = 6;
 };
 
 message StringIntLabelMap {

research/object_detection/samples/configs/ssdlite_mobiledet_cpu_320x320_coco_sync_4x4.config

+# SSDLite with MobileDet-CPU feature extractor.
+# Reference: Xiong & Liu et al., https://arxiv.org/abs/2004.14525
+# Trained on COCO, initialized from scratch.
+#
+# 0.45B MulAdds, 4.21M Parameters. Latency is 113ms on Pixel 1 CPU.
+# Achieves 24.1 mAP on COCO14 minival dataset.
+# Achieves 23.5 mAP on COCO17 val dataset.
+#
+# This config is TPU compatible.
+
+model {
+  ssd {
+    inplace_batchnorm_update: true
+    freeze_batchnorm: false
+    num_classes: 90
+    box_coder {
+      faster_rcnn_box_coder {
+        y_scale: 10.0
+        x_scale: 10.0
+        height_scale: 5.0
+        width_scale: 5.0
+      }
+    }
+    matcher {
+      argmax_matcher {
+        matched_threshold: 0.5
+        unmatched_threshold: 0.5
+        ignore_thresholds: false
+        negatives_lower_than_unmatched: true
+        force_match_for_each_row: true
+        use_matmul_gather: true
+      }
+    }
+    similarity_calculator {
+      iou_similarity {
+      }
+    }
+    encode_background_as_zeros: true
+    anchor_generator {
+      ssd_anchor_generator {
+        num_layers: 6
+        min_scale: 0.2
+        max_scale: 0.95
+        aspect_ratios: 1.0
+        aspect_ratios: 2.0
+        aspect_ratios: 0.5
+        aspect_ratios: 3.0
+        aspect_ratios: 0.3333
+      }
+    }
+    image_resizer {
+      fixed_shape_resizer {
+        height: 320
+        width: 320
+      }
+    }
+    box_predictor {
+      convolutional_box_predictor {
+        min_depth: 0
+        max_depth: 0
+        num_layers_before_predictor: 0
+        use_dropout: false
+        dropout_keep_probability: 0.8
+        kernel_size: 3
+        use_depthwise: true
+        box_code_size: 4
+        apply_sigmoid_to_scores: false
+        class_prediction_bias_init: -4.6
+        conv_hyperparams {
+          activation: RELU_6,
+          regularizer {
+            l2_regularizer {
+              weight: 0.00004
+            }
+          }
+          initializer {
+            random_normal_initializer {
+              stddev: 0.03
+              mean: 0.0
+            }
+          }
+          batch_norm {
+            train: true,
+            scale: true,
+            center: true,
+            decay: 0.97,
+            epsilon: 0.001,
+          }
+        }
+      }
+    }
+    feature_extractor {
+      type: 'ssd_mobiledet_cpu'
+      min_depth: 16
+      depth_multiplier: 1.0
+      use_depthwise: true
+      conv_hyperparams {
+        activation: RELU_6,
+        regularizer {
+          l2_regularizer {
+            weight: 0.00004
+          }
+        }
+        initializer {
+          truncated_normal_initializer {
+            stddev: 0.03
+            mean: 0.0
+          }
+        }
+        batch_norm {
+          train: true,
+          scale: true,
+          center: true,
+          decay: 0.97,
+          epsilon: 0.001,
+        }
+      }
+      override_base_feature_extractor_hyperparams: false
+    }
+    loss {
+      classification_loss {
+        weighted_sigmoid_focal {
+          alpha: 0.75,
+          gamma: 2.0
+        }
+      }
+      localization_loss {
+        weighted_smooth_l1 {
+          delta: 1.0
+        }
+      }
+      classification_weight: 1.0
+      localization_weight: 1.0
+    }
+    normalize_loss_by_num_matches: true
+    normalize_loc_loss_by_codesize: true
+    post_processing {
+      batch_non_max_suppression {
+        score_threshold: 1e-8
+        iou_threshold: 0.6
+        max_detections_per_class: 100
+        max_total_detections: 100
+        use_static_shapes: true
+      }
+      score_converter: SIGMOID
+    }
+  }
+}
+
+train_config: {
+  batch_size: 512
+  sync_replicas: true
+  startup_delay_steps: 0
+  replicas_to_aggregate: 32
+  num_steps: 400000
+  data_augmentation_options {
+    random_horizontal_flip {
+    }
+  }
+  data_augmentation_options {
+    ssd_random_crop {
+    }
+  }
+  optimizer {
+    momentum_optimizer: {
+      learning_rate: {
+        cosine_decay_learning_rate {
+          learning_rate_base: 0.8
+          total_steps: 400000
+          warmup_learning_rate: 0.13333
+          warmup_steps: 2000
+        }
+      }
+      momentum_optimizer_value: 0.9
+    }
+    use_moving_average: false
+  }
+  max_number_of_boxes: 100
+  unpad_groundtruth_tensors: false
+}
+
+train_input_reader: {
+  label_map_path: "PATH_TO_BE_CONFIGURED/mscoco_label_map.pbtxt"
+  tf_record_input_reader {
+    input_path: "PATH_TO_BE_CONFIGURED/mscoco_train.record-?????-of-00100"
+  }
+}
+
+eval_config: {
+  metrics_set: "coco_detection_metrics"
+  use_moving_averages: false
+  num_examples: 8000
+}
+
+eval_input_reader: {
+  label_map_path: "PATH_TO_BE_CONFIGURED/mscoco_label_map.pbtxt"
+  shuffle: false
+  num_epochs: 1
+  tf_record_input_reader {
+    input_path: "PATH_TO_BE_CONFIGURED/mscoco_val.record-?????-of-00010"
+  }
+}

research/object_detection/samples/configs/ssdlite_mobiledet_dsp_320x320_coco_sync_4x4.config

+# SSDLite with MobileDet-DSP feature extractor.
+# Reference: Xiong & Liu et al., https://arxiv.org/abs/2004.14525
+# Trained on COCO, initialized from scratch.
+#
+# 2.82B MulAdds, 7.16M Parameters. Latency is 12.3ms on Pixel 4 DSP.
+# Achieves 28.9 mAP on COCO14 minival dataset.
+# Achieves 28.4 mAP on COCO17 val dataset.
+#
+# This config is TPU compatible.
+
+model {
+  ssd {
+    inplace_batchnorm_update: true
+    freeze_batchnorm: false
+    num_classes: 90
+    box_coder {
+      faster_rcnn_box_coder {
+        y_scale: 10.0
+        x_scale: 10.0
+        height_scale: 5.0
+        width_scale: 5.0
+      }
+    }
+    matcher {
+      argmax_matcher {
+        matched_threshold: 0.5
+        unmatched_threshold: 0.5
+        ignore_thresholds: false
+        negatives_lower_than_unmatched: true
+        force_match_for_each_row: true
+        use_matmul_gather: true
+      }
+    }
+    similarity_calculator {
+      iou_similarity {
+      }
+    }
+    encode_background_as_zeros: true
+    anchor_generator {
+      ssd_anchor_generator {
+        num_layers: 6
+        min_scale: 0.2
+        max_scale: 0.95
+        aspect_ratios: 1.0
+        aspect_ratios: 2.0
+        aspect_ratios: 0.5
+        aspect_ratios: 3.0
+        aspect_ratios: 0.3333
+      }
+    }
+    image_resizer {
+      fixed_shape_resizer {
+        height: 320
+        width: 320
+      }
+    }
+    box_predictor {
+      convolutional_box_predictor {
+        min_depth: 0
+        max_depth: 0
+        num_layers_before_predictor: 0
+        use_dropout: false
+        dropout_keep_probability: 0.8
+        kernel_size: 3
+        use_depthwise: true
+        box_code_size: 4
+        apply_sigmoid_to_scores: false
+        class_prediction_bias_init: -4.6
+        conv_hyperparams {
+          activation: RELU_6,
+          regularizer {
+            l2_regularizer {
+              weight: 0.00004
+            }
+          }
+          initializer {
+            random_normal_initializer {
+              stddev: 0.03
+              mean: 0.0
+            }
+          }
+          batch_norm {
+            train: true,
+            scale: true,
+            center: true,
+            decay: 0.97,
+            epsilon: 0.001,
+          }
+        }
+      }
+    }
+    feature_extractor {
+      type: 'ssd_mobiledet_dsp'
+      min_depth: 16
+      depth_multiplier: 1.0
+      use_depthwise: true
+      conv_hyperparams {
+        activation: RELU_6,
+        regularizer {
+          l2_regularizer {
+            weight: 0.00004
+          }
+        }
+        initializer {
+          truncated_normal_initializer {
+            stddev: 0.03
+            mean: 0.0
+          }
+        }
+        batch_norm {
+          train: true,
+          scale: true,
+          center: true,
+          decay: 0.97,
+          epsilon: 0.001,
+        }
+      }
+      override_base_feature_extractor_hyperparams: false
+    }
+    loss {
+      classification_loss {
+        weighted_sigmoid_focal {
+          alpha: 0.75,
+          gamma: 2.0
+        }
+      }
+      localization_loss {
+        weighted_smooth_l1 {
+          delta: 1.0
+        }
+      }
+      classification_weight: 1.0
+      localization_weight: 1.0
+    }
+    normalize_loss_by_num_matches: true
+    normalize_loc_loss_by_codesize: true
+    post_processing {
+      batch_non_max_suppression {
+        score_threshold: 1e-8
+        iou_threshold: 0.6
+        max_detections_per_class: 100
+        max_total_detections: 100
+        use_static_shapes: true
+      }
+      score_converter: SIGMOID
+    }
+  }
+}
+
+train_config: {
+  batch_size: 512
+  sync_replicas: true
+  startup_delay_steps: 0
+  replicas_to_aggregate: 32
+  num_steps: 400000
+  data_augmentation_options {
+    random_horizontal_flip {
+    }
+  }
+  data_augmentation_options {
+    ssd_random_crop {
+    }
+  }
+  optimizer {
+    momentum_optimizer: {
+      learning_rate: {
+        cosine_decay_learning_rate {
+          learning_rate_base: 0.8
+          total_steps: 400000
+          warmup_learning_rate: 0.13333
+          warmup_steps: 2000
+        }
+      }
+      momentum_optimizer_value: 0.9
+    }
+    use_moving_average: false
+  }
+  max_number_of_boxes: 100
+  unpad_groundtruth_tensors: false
+}
+
+train_input_reader: {
+  label_map_path: "PATH_TO_BE_CONFIGURED/mscoco_label_map.pbtxt"
+  tf_record_input_reader {
+    input_path: "PATH_TO_BE_CONFIGURED/mscoco_train.record-?????-of-00100"
+  }
+}
+
+eval_config: {
+  metrics_set: "coco_detection_metrics"
+  use_moving_averages: false
+  num_examples: 8000
+}
+
+eval_input_reader: {
+  label_map_path: "PATH_TO_BE_CONFIGURED/mscoco_label_map.pbtxt"
+  shuffle: false
+  num_epochs: 1
+  tf_record_input_reader {
+    input_path: "PATH_TO_BE_CONFIGURED/mscoco_val.record-?????-of-00010"
+  }
+}

research/object_detection/samples/configs/ssdlite_mobiledet_edgetpu_320x320_coco_sync_4x4.config

+# SSDLite with MobileDet-EdgeTPU feature extractor.
+# Reference: Xiong & Liu et al., https://arxiv.org/abs/2004.14525
+# Trained on COCO, initialized from scratch.
+#
+# 1.53B MulAdds, 4.20M Parameters. Latency is 6.9ms on Pixel 4 EdgeTPU.
+# Achieves 26.0 mAP on COCO14 minival dataset.
+# Achieves 25.2 mAP on COCO17 val dataset.
+#
+# This config is TPU compatible.
+
+model {
+  ssd {
+    inplace_batchnorm_update: true
+    freeze_batchnorm: false
+    num_classes: 90
+    box_coder {
+      faster_rcnn_box_coder {
+        y_scale: 10.0
+        x_scale: 10.0
+        height_scale: 5.0
+        width_scale: 5.0
+      }
+    }
+    matcher {
+      argmax_matcher {
+        matched_threshold: 0.5
+        unmatched_threshold: 0.5
+        ignore_thresholds: false
+        negatives_lower_than_unmatched: true
+        force_match_for_each_row: true
+        use_matmul_gather: true
+      }
+    }
+    similarity_calculator {
+      iou_similarity {
+      }
+    }
+    encode_background_as_zeros: true
+    anchor_generator {
+      ssd_anchor_generator {
+        num_layers: 6
+        min_scale: 0.2
+        max_scale: 0.95
+        aspect_ratios: 1.0
+        aspect_ratios: 2.0
+        aspect_ratios: 0.5
+        aspect_ratios: 3.0
+        aspect_ratios: 0.3333
+      }
+    }
+    image_resizer {
+      fixed_shape_resizer {
+        height: 320
+        width: 320
+      }
+    }
+    box_predictor {
+      convolutional_box_predictor {
+        min_depth: 0
+        max_depth: 0
+        num_layers_before_predictor: 0
+        use_dropout: false
+        dropout_keep_probability: 0.8
+        kernel_size: 3
+        use_depthwise: true
+        box_code_size: 4
+        apply_sigmoid_to_scores: false
+        class_prediction_bias_init: -4.6
+        conv_hyperparams {
+          activation: RELU_6,
+          regularizer {
+            l2_regularizer {
+              weight: 0.00004
+            }
+          }
+          initializer {
+            random_normal_initializer {
+              stddev: 0.03
+              mean: 0.0
+            }
+          }
+          batch_norm {
+            train: true,
+            scale: true,
+            center: true,
+            decay: 0.97,
+            epsilon: 0.001,
+          }
+        }
+      }
+    }
+    feature_extractor {
+      type: 'ssd_mobiledet_edgetpu'
+      min_depth: 16
+      depth_multiplier: 1.0
+      use_depthwise: true
+      conv_hyperparams {
+        activation: RELU_6,
+        regularizer {
+          l2_regularizer {
+            weight: 0.00004
+          }
+        }
+        initializer {
+          truncated_normal_initializer {
+            stddev: 0.03
+            mean: 0.0
+          }
+        }
+        batch_norm {
+          train: true,
+          scale: true,
+          center: true,
+          decay: 0.97,
+          epsilon: 0.001,
+        }
+      }
+      override_base_feature_extractor_hyperparams: false
+    }
+    loss {
+      classification_loss {
+        weighted_sigmoid_focal {
+          alpha: 0.75,
+          gamma: 2.0
+        }
+      }
+      localization_loss {
+        weighted_smooth_l1 {
+          delta: 1.0
+        }
+      }
+      classification_weight: 1.0
+      localization_weight: 1.0
+    }
+    normalize_loss_by_num_matches: true
+    normalize_loc_loss_by_codesize: true
+    post_processing {
+      batch_non_max_suppression {
+        score_threshold: 1e-8
+        iou_threshold: 0.6
+        max_detections_per_class: 100
+        max_total_detections: 100
+        use_static_shapes: true
+      }
+      score_converter: SIGMOID
+    }
+  }
+}
+
+train_config: {
+  batch_size: 512
+  sync_replicas: true
+  startup_delay_steps: 0
+  replicas_to_aggregate: 32
+  num_steps: 400000
+  data_augmentation_options {
+    random_horizontal_flip {
+    }
+  }
+  data_augmentation_options {
+    ssd_random_crop {
+    }
+  }
+  optimizer {
+    momentum_optimizer: {
+      learning_rate: {
+        cosine_decay_learning_rate {
+          learning_rate_base: 0.8
+          total_steps: 400000
+          warmup_learning_rate: 0.13333
+          warmup_steps: 2000
+        }
+      }
+      momentum_optimizer_value: 0.9
+    }
+    use_moving_average: false
+  }
+  max_number_of_boxes: 100
+  unpad_groundtruth_tensors: false
+}
+
+train_input_reader: {
+  label_map_path: "PATH_TO_BE_CONFIGURED/mscoco_label_map.pbtxt"
+  tf_record_input_reader {
+    input_path: "PATH_TO_BE_CONFIGURED/mscoco_train.record-?????-of-00100"
+  }
+}
+
+eval_config: {
+  metrics_set: "coco_detection_metrics"
+  use_moving_averages: false
+  num_examples: 8000
+}
+
+eval_input_reader: {
+  label_map_path: "PATH_TO_BE_CONFIGURED/mscoco_label_map.pbtxt"
+  shuffle: false
+  num_epochs: 1
+  tf_record_input_reader {
+    input_path: "PATH_TO_BE_CONFIGURED/mscoco_val.record-?????-of-00010"
+  }
+}

research/object_detection/test_data/context_rcnn_camera_trap.config

+
+# Context R-CNN unit test configuration using sequence example sample data with
+# context_features created from the Snapshot Serengeti Dataset and stored in
+# snapshot_serengeti_sequence_examples.record.
+# This model uses attention into contextual features within the Faster R-CNN
+# object detection framework to improve object detection performance.
+# See https://arxiv.org/abs/1912.03538 for more information.
+
+model {
+  faster_rcnn {
+    num_classes: 48
+    image_resizer {
+      fixed_shape_resizer {
+        height: 640
+        width: 640
+      }
+    }
+    feature_extractor {
+      type: "faster_rcnn_resnet101"
+      first_stage_features_stride: 16
+      batch_norm_trainable: true
+    }
+    first_stage_anchor_generator {
+      grid_anchor_generator {
+        height_stride: 16
+        width_stride: 16
+        scales: 0.25
+        scales: 0.5
+        scales: 1.0
+        scales: 2.0
+        aspect_ratios: 0.5
+        aspect_ratios: 1.0
+        aspect_ratios: 2.0
+      }
+    }
+    first_stage_box_predictor_conv_hyperparams {
+      op: CONV
+      regularizer {
+        l2_regularizer {
+          weight: 0.0
+        }
+      }
+      initializer {
+        truncated_normal_initializer {
+          stddev: 0.00999999977648
+        }
+      }
+    }
+    first_stage_nms_score_threshold: 0.0
+    first_stage_nms_iou_threshold: 0.699999988079
+    first_stage_max_proposals: 300
+    first_stage_localization_loss_weight: 2.0
+    first_stage_objectness_loss_weight: 1.0
+    initial_crop_size: 14
+    maxpool_kernel_size: 2
+    maxpool_stride: 2
+    second_stage_box_predictor {
+      mask_rcnn_box_predictor {
+        fc_hyperparams {
+          op: FC
+          regularizer {
+            l2_regularizer {
+              weight: 0.0
+            }
+          }
+          initializer {
+            variance_scaling_initializer {
+              factor: 1.0
+              uniform: true
+              mode: FAN_AVG
+            }
+          }
+        }
+        use_dropout: false
+        dropout_keep_probability: 1.0
+        share_box_across_classes: true
+      }
+    }
+    second_stage_post_processing {
+      batch_non_max_suppression {
+        score_threshold: 0.0
+        iou_threshold: 0.600000023842
+        max_detections_per_class: 100
+        max_total_detections: 300
+      }
+      score_converter: SOFTMAX
+    }
+    second_stage_localization_loss_weight: 2.0
+    second_stage_classification_loss_weight: 1.0
+    use_matmul_crop_and_resize: true
+    clip_anchors_to_image: true
+    use_matmul_gather_in_matcher: true
+    use_static_balanced_label_sampler: true
+    use_static_shapes: true
+    context_config {
+      max_num_context_features: 2000
+      context_feature_length: 2057
+    }
+  }
+}
+train_config {
+  batch_size: 1
+  data_augmentation_options {
+    random_horizontal_flip {
+    }
+  }
+  sync_replicas: true
+  optimizer {
+    momentum_optimizer {
+      learning_rate {
+        manual_step_learning_rate {
+          initial_learning_rate: 0.0
+          schedule {
+            step: 2000
+            learning_rate: 0.00200000009499
+          }
+          schedule {
+            step: 200000
+            learning_rate: 0.000199999994948
+          }
+          schedule {
+            step: 300000
+            learning_rate: 1.99999994948e-05
+          }
+          warmup: true
+        }
+      }
+      momentum_optimizer_value: 0.899999976158
+    }
+    use_moving_average: false
+  }
+  gradient_clipping_by_norm: 10.0
+  num_steps: 500000
+  replicas_to_aggregate: 8
+  max_number_of_boxes: 100
+  unpad_groundtruth_tensors: false
+}
+train_input_reader {
+  label_map_path: ""
+  tf_record_input_reader {
+    input_path: ""
+  }
+  load_context_features: true
+  input_type: TF_SEQUENCE_EXAMPLE
+}
+eval_config {
+  max_evals: 50
+  metrics_set: "coco_detection_metrics"
+  use_moving_averages: false
+  batch_size: 1
+}
+eval_input_reader {
+  label_map_path: ""
+  shuffle: false
+  num_epochs: 1
+  tf_record_input_reader {
+    input_path: ""
+  }
+  load_context_features: true
+  input_type: TF_SEQUENCE_EXAMPLE
+}

research/object_detection/tpu_exporters/export_saved_model_tpu.py

@@ -17,7 +17,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 from object_detection.tpu_exporters import export_saved_model_tpu_lib
 
 flags = tf.app.flags

research/object_detection/tpu_exporters/export_saved_model_tpu_lib.py

@@ -17,7 +17,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 from google.protobuf import text_format
 # pylint: disable=g-direct-tensorflow-import

research/object_detection/tpu_exporters/export_saved_model_tpu_lib_test.py

@@ -22,7 +22,7 @@ import os
 
 from absl.testing import parameterized
 import numpy as np
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 from object_detection.tpu_exporters import export_saved_model_tpu_lib
 

research/object_detection/tpu_exporters/faster_rcnn.py

@@ -18,7 +18,7 @@ from __future__ import division
 from __future__ import print_function
 
 # pylint: disable=protected-access
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 # pylint: disable=g-import-not-at-top
 # Checking TF version, because this module relies on TPUPartitionedCall

research/object_detection/tpu_exporters/ssd.py

@@ -17,7 +17,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 # pylint: disable=g-import-not-at-top
 # Checking TF version, because this module relies on TPUPartitionedCall

research/object_detection/tpu_exporters/utils.py

@@ -17,7 +17,7 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 
 def bfloat16_to_float32(tensor):

research/object_detection/tpu_exporters/utils_test.py

@@ -20,7 +20,7 @@ from __future__ import division
 from __future__ import print_function
 
 from six.moves import range
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 from object_detection.tpu_exporters import utils
 

research/object_detection/utils/autoaugment_utils.py

@@ -22,7 +22,7 @@ from __future__ import print_function
 import inspect
 import math
 import six
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 # pylint: disable=g-import-not-at-top
 try:

research/object_detection/utils/category_util.py

@@ -21,7 +21,7 @@ from __future__ import print_function
 
 import csv
 
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 
 def load_categories_from_csv_file(csv_path):

research/object_detection/utils/category_util_test.py

@@ -21,7 +21,7 @@ from __future__ import print_function
 
 import os
 
-import tensorflow as tf
+import tensorflow.compat.v1 as tf
 
 from object_detection.utils import category_util