tensorflow/models

# Faster R-CNN with Inception Resnet v2, Atrous version;
# Configured for Open Images Dataset.
# Users should configure the fine_tune_checkpoint field in the train config as
# well as the label_map_path and input_path fields in the train_input_reader and
# eval_input_reader. Search for "PATH_TO_BE_CONFIGURED" to find the fields that
# should be configured.

model {
  faster_rcnn {
    num_classes: 546
    image_resizer {
      keep_aspect_ratio_resizer {
        min_dimension: 600
        max_dimension: 1024
      }
    }
    feature_extractor {
      type: 'faster_rcnn_inception_resnet_v2'
      first_stage_features_stride: 8
    }
    first_stage_anchor_generator {
      grid_anchor_generator {
        scales: [0.25, 0.5, 1.0, 2.0]
        aspect_ratios: [0.5, 1.0, 2.0]
        height_stride: 8
        width_stride: 8
      }
    }
    first_stage_atrous_rate: 2
    first_stage_box_predictor_conv_hyperparams {
      op: CONV
      regularizer {
        l2_regularizer {
          weight: 0.0
        }
      }
      initializer {
        truncated_normal_initializer {
          stddev: 0.01
        }
      }
    }
    first_stage_nms_score_threshold: 0.0
    first_stage_nms_iou_threshold: 0.7
    first_stage_max_proposals: 300
    first_stage_localization_loss_weight: 2.0
    first_stage_objectness_loss_weight: 1.0
    initial_crop_size: 17
    maxpool_kernel_size: 1
    maxpool_stride: 1
    second_stage_box_predictor {
      mask_rcnn_box_predictor {
        use_dropout: false
        dropout_keep_probability: 1.0
        fc_hyperparams {
          op: FC
          regularizer {
            l2_regularizer {
              weight: 0.0
            }
          }
          initializer {
            variance_scaling_initializer {
              factor: 1.0
              uniform: true
              mode: FAN_AVG
            }
          }
        }
      }
    }
    second_stage_post_processing {
      batch_non_max_suppression {
        score_threshold: 0.0
        iou_threshold: 0.6
        max_detections_per_class: 100
        max_total_detections: 100
      }
      score_converter: SOFTMAX
    }
    second_stage_localization_loss_weight: 2.0
    second_stage_classification_loss_weight: 1.0
  }
}

train_config: {
  batch_size: 1
  optimizer {
    momentum_optimizer: {
      learning_rate: {
        manual_step_learning_rate {
          initial_learning_rate: 0.00006
          schedule {
            step: 6000000
            learning_rate: .000006
          }
          schedule {
            step: 7000000
            learning_rate: .0000006
          }
        }
      }
      momentum_optimizer_value: 0.9
    }
    use_moving_average: false
  }
  gradient_clipping_by_norm: 10.0
  fine_tune_checkpoint: "PATH_TO_BE_CONFIGURED/model.ckpt"
  # Note: The below line limits the training process to 800K steps, which we
  # empirically found to be sufficient enough to train the Open Images dataset.
  # This effectively bypasses the learning rate schedule (the learning rate will
  # never decay). Remove the below line to train indefinitely.
  num_steps: 8000000
  data_augmentation_options {
    random_horizontal_flip {
    }
  }
}

train_input_reader: {
  tf_record_input_reader {
    input_path: "PATH_TO_BE_CONFIGURED/oid_bbox_trainable_train.record"
  }
  label_map_path: "PATH_TO_BE_CONFIGURED/oid_bbox_trainable_label_map.pbtxt"
}

eval_config: {
  # Note: this metric is deprecated, use oid_challenge_detection_metrics.
  metrics_set: "open_images_metrics"
  num_examples: 8000
  # Note: The below line limits the evaluation process to 10 evaluations.
  # Remove the below line to evaluate indefinitely.
  max_evals: 10
}

eval_input_reader: {
  tf_record_input_reader {
    input_path: "PATH_TO_BE_CONFIGURED/oid_bbox_trainable_val.record"
  }
  label_map_path: "PATH_TO_BE_CONFIGURED/oid_bbox_trainable_label_map.pbtxt"
  shuffle: false
  num_readers: 1
}