package protos

Get desktop application:
View/edit binary Protocol Buffers messages

Used in: Optimizer

• optional LearningRate learning_rate = 1

• optional float rho = 2

Configuration message for the AdamOptimizer See: https://www.tensorflow.org/api_docs/python/tf/train/AdamOptimizer

Used in: Optimizer

• optional LearningRate learning_rate = 1

Used in: Predictor

• optional RnnCell rnn_cell = 1

• optional Regularizer rnn_regularizer = 2

• optional int32 num_attention_units = 3

• optional int32 max_num_steps = 4

• optional bool multi_attention = 5

• optional int32 beam_width = 6

• optional bool reverse = 7

• optional LabelMap label_map = 8

• optional Loss loss = 9

• optional LanguageModelRnnCell lm_rnn_cell = 10

• optional bool sync = 11

Configuration proto for batch norm to apply after convolution op. See https://www.tensorflow.org/api_docs/python/tf/contrib/layers/batch_norm

Used in: Hyperparams

• optional float decay = 1

• optional bool center = 2

• optional bool scale = 3

• optional float epsilon = 4

• optional bool train = 5

  Whether to train the batch norm variables. If this is set to false during training, the current value of the batch_norm variables are used for forward pass but they are never updated.

Used in: FeatureExtractor, TextGenerator

• optional bool static = 1

• optional RnnCell fw_bw_rnn_cell = 2

• optional Regularizer rnn_regularizer = 3

• optional int32 num_output_units = 4

• optional Hyperparams fc_hyperparams = 5

• optional bool summarize_activations = 6

Used in: LabelMap, StringFiltering

• oneof source_oneof

  • string text_file = 1

  • string text_string = 2

  • CharacterSet.BuiltInSet built_in_set = 3

Used in: CharacterSet

• LOWERCASE = 0

• ALLCASES = 1

• ALLCASES_SYMBOLS = 2

Configuration message for a constant learning rate.

Used in: LearningRate

• optional float learning_rate = 1

Used in: FeatureExtractor, SpatialTransformer, TextGenerator

• oneof convnet_oneof

  • CrnnNet crnn_net = 1

  • ResNet resnet = 2

  • StnConvnet stn_convnet = 3

  • TgnDeconvnet tgn_deconvnet = 4

Used in: Convnet

• optional CrnnNet.NetType net_type = 1

• optional Hyperparams conv_hyperparams = 2

• optional bool summarize_activations = 3

• optional bool tiny = 4

Used in: CrnnNet

• SINGLE_BRANCH = 0

• TWO_BRANCHES = 1

• THREE_BRANCHES = 2

Used in: Predictor

(message has no fields)

Message for configuring DetectionModel evaluation jobs (eval.py).

Used in: TrainEvalPipelineConfig

• optional uint32 num_visualizations = 1

  Number of visualization images to generate.

• optional bool only_visualize_incorrect = 14

• optional uint32 num_examples = 2

  Number of examples to process of evaluation.

• optional uint32 eval_interval_secs = 3

  How often to run evaluation.

• optional uint32 max_evals = 4

  Maximum number of times to run evaluation. If set to 0, will run forever.

• optional bool save_graph = 5

  Whether the TensorFlow graph used for evaluation should be saved to disk.

• optional string visualization_export_dir = 6

  Path to directory to store visualizations in. If empty, visualization images are not exported (only shown on Tensorboard).

• optional string eval_master = 7

  BNS name of the TensorFlow master.

• optional string metrics_set = 8

  Type of metrics to use for evaluation. Currently supports only Pascal VOC detection metrics.

• optional string export_path = 9

  Path to export detections to COCO compatible JSON format.

• optional bool ignore_groundtruth = 10

  Option to not read groundtruth labels and only export detections to COCO-compatible JSON file.

• optional bool use_moving_averages = 11

  Use exponential moving averages of variables for evaluation. TODO: When this is false make sure the model is constructed without moving averages in restore_fn.

• optional bool eval_instance_masks = 12

  Whether to evaluate instance masks.

• optional bool eval_with_lexicon = 15

  Whether to evaluate with lexicon

• repeated PreprocessingStep data_preprocessing_steps = 13

  data preprocessing steps

Configuration message for an exponentially decaying learning rate. See https://www.tensorflow.org/versions/master/api_docs/python/train/ \ decaying_the_learning_rate#exponential_decay

Used in: LearningRate

• optional float initial_learning_rate = 1

• optional uint32 decay_steps = 2

• optional float decay_factor = 3

• optional bool staircase = 4

An externally defined input reader. Users may define an extension to this proto to interface their own input readers.

Used in: InputReader

(message has no fields)

Used in: MultiPredictorsRecognitionModel

• optional Convnet convnet = 1

• repeated BidirectionalRnn bidirectional_rnn = 2

• optional bool summarize_activations = 3

Used in: RnnCell

• optional uint32 num_units = 1

• optional Initializer initializer = 2

Configuration proto for the convolution op hyperparameters to use in the object detection pipeline.

Used in: BidirectionalRnn, CrnnNet, ResNet, SpatialTransformer, StnConvnet, TgnDeconvnet

• optional Hyperparams.Op op = 1

• optional Regularizer regularizer = 2

  Regularizer for the weights of the convolution op.

• optional Initializer initializer = 3

  Initializer for the weights of the convolution op.

• optional Hyperparams.Activation activation = 4

• optional BatchNorm batch_norm = 5

  BatchNorm hyperparameters. If this parameter is NOT set then BatchNorm is not applied!

Type of activation to apply after convolution.

Used in: Hyperparams

• NONE = 0

  Use None (no activation)

• RELU = 1

  Use tf.nn.relu

• RELU_6 = 2

  Use tf.nn.relu6

Operations affected by hyperparameters.

Used in: Hyperparams

• CONV = 1

  Convolution, Separable Convolution, Convolution transpose.

• FC = 2

  Fully connected

Used in: PreprocessingStep

(message has no fields)

Proto with one-of field for initializers.

Used in: GruCell, Hyperparams, LstmCell

• oneof initializer_oneof

  • TruncatedNormalInitializer truncated_normal_initializer = 1

  • VarianceScalingInitializer variance_scaling_initializer = 2

  • OrthogonalInitializer orthogonal_initializer = 3

  • UniformInitializer uniform_initializer = 4

Used in: TrainEvalPipelineConfig

• optional string label_map_path = 1

  Path to StringIntLabelMap pbtxt file specifying the mapping from string labels to integer ids.

• optional bool shuffle = 2

  Whether data should be processed in the order they are read in, or shuffled randomly.

• optional uint32 queue_capacity = 3

  Maximum number of records to keep in reader queue.

• optional uint32 min_after_dequeue = 4

  Minimum number of records to keep in reader queue. A large value is needed to generate a good random shuffle.

• optional uint32 num_epochs = 5

  The number of times a data source is read. If set to zero, the data source will be reused indefinitely.

• optional uint32 num_readers = 6

  Number of reader instances to create.

• optional bool load_instance_masks = 7

  Whether to load groundtruth instance masks.

• oneof input_reader

  • TFRecordInputReader tf_record_input_reader = 8

  • ExternalInputReader external_input_reader = 9

Used in: Loss

• optional float weight = 3

Configuration proto for L1 Regularizer. See https://www.tensorflow.org/api_docs/python/tf/contrib/layers/l1_regularizer

Used in: Regularizer

• optional float weight = 1

Used in: Loss

• optional float weight = 3

Configuration proto for L2 Regularizer. See https://www.tensorflow.org/api_docs/python/tf/contrib/layers/l2_regularizer

Used in: Regularizer

• optional float weight = 1

Used in: AttentionPredictor

• optional CharacterSet character_set = 1

• optional int64 label_offset = 2

• optional int64 unk_label = 3

Used in: AttentionPredictor

• repeated RnnCell rnn_cell = 1

• optional string restore_path = 2

Configuration message for optimizer learning rate.

Used in: AdadeltaOptimizer, AdamOptimizer, MomentumOptimizer, NadamOptimizer, RMSPropOptimizer

• oneof learning_rate

  • ConstantLearningRate constant_learning_rate = 1

  • ExponentialDecayLearningRate exponential_decay_learning_rate = 2

  • ManualStepLearningRate manual_step_learning_rate = 3

Used in: AttentionPredictor, MultiPredictorsRecognitionModel, TextGenerator

• oneof loss_oneof

  • SequenceCrossEntropyLoss sequence_cross_entropy_loss = 1

  • L2RegressionLoss l2_regression_loss = 2

  • L1RegressionLoss l1_regression_loss = 3

Used in: RnnCell

• optional uint32 num_units = 1

• optional bool use_peepholes = 2

• optional float forget_bias = 3

• optional Initializer initializer = 4

Configuration message for a manually defined learning rate schedule.

Used in: LearningRate

• optional float initial_learning_rate = 1

• repeated ManualStepLearningRate.LearningRateSchedule schedule = 2

Used in: ManualStepLearningRate

• optional uint32 step = 1

• optional float learning_rate = 2

Used in: TrainEvalPipelineConfig

• oneof model_oneof

  • MultiPredictorsRecognitionModel multi_predictors_recognition_model = 1

Configuration message for the MomentumOptimizer See: https://www.tensorflow.org/api_docs/python/tf/train/MomentumOptimizer

Used in: Optimizer

• optional LearningRate learning_rate = 1

• optional float momentum_optimizer_value = 2

Used in: Model

• optional SpatialTransformer spatial_transformer = 1

• optional FeatureExtractor feature_extractor = 2

• repeated Predictor predictor = 3

• optional bool keypoint_supervision = 4

• optional Loss regression_loss = 5

• optional TextGenerator text_generator = 6

Used in: Optimizer

• optional LearningRate learning_rate = 1

Used in: PreprocessingStep

• optional float original_minval = 1

• optional float original_maxval = 2

• optional float target_minval = 3

• optional float target_maxval = 4

Top level optimizer message.

Used in: TrainConfig

• oneof optimizer

  • RMSPropOptimizer rms_prop_optimizer = 1

  • MomentumOptimizer momentum_optimizer = 2

  • AdamOptimizer adam_optimizer = 3

  • NadamOptimizer nadam_optimizer = 4

  • AdadeltaOptimizer adadelta_optimizer = 5

• optional bool use_moving_average = 6

• optional float moving_average_decay = 7

Used in: Initializer

• optional float gain = 1

• optional int32 seed = 2

Used in: MultiPredictorsRecognitionModel

• optional string name = 1

• oneof predictor_oneof

  • AttentionPredictor attention_predictor = 2

  • CtcPredictor ctc_predictor = 3

Used in: EvalConfig, TrainConfig

• oneof preprocessing_step

  • ResizeImageRandomMethod resize_image_random_method = 1

  • ResizeImage resize_image = 2

  • NormalizeImage normalize_image = 3

  • RandomPixelValueScale random_pixel_value_scale = 4

  • RandomRgbToGray random_rgb_to_gray = 5

  • RandomAdjustBrightness random_adjust_brightness = 6

  • RandomAdjustContrast random_adjust_contrast = 7

  • RandomAdjustHue random_adjust_hue = 8

  • RandomAdjustSaturation random_adjust_saturation = 9

  • RandomDistortColor random_distort_color = 10

  • ImageToFloat image_to_float = 11

  • SubtractChannelMean subtract_channel_mean = 12

  • RgbToGray rgb_to_gray = 13

  • StringFiltering string_filtering = 14

Configuration message for the RMSPropOptimizer See: https://www.tensorflow.org/api_docs/python/tf/train/RMSPropOptimizer

Used in: Optimizer

• optional LearningRate learning_rate = 1

• optional float momentum_optimizer_value = 2

• optional float decay = 3

• optional float epsilon = 4

Used in: PreprocessingStep

• optional float max_delta = 1

Used in: PreprocessingStep

• optional float min_delta = 1

• optional float max_delta = 2

Used in: PreprocessingStep

• optional float max_delta = 1

Used in: PreprocessingStep

• optional float min_delta = 1

• optional float max_delta = 2

Used in: PreprocessingStep

• optional int32 color_ordering = 1

Used in: PreprocessingStep

• optional float minval = 1

• optional float maxval = 2

Used in: PreprocessingStep

• optional float probability = 1

Proto with one-of field for regularizers.

Used in: AttentionPredictor, BidirectionalRnn, Hyperparams

• oneof regularizer_oneof

  • L1Regularizer l1_regularizer = 1

  • L2Regularizer l2_regularizer = 2

Used in: Convnet

• optional ResNet.NetType net_type = 1

• optional ResNet.NetDepth net_depth = 2

• optional Hyperparams conv_hyperparams = 3

• optional bool summarize_activations = 4

Used in: ResNet

• RESNET_30 = 0

• RESNET_50 = 1

• RESNET_100 = 2

Used in: ResNet

• SINGLE_BRANCH = 0

• TWO_BRANCHES = 1

• THREE_BRANCHES = 2

Used in: PreprocessingStep

• optional int32 target_height = 1

• optional int32 target_width = 2

• optional ResizeImage.Method method = 3

Used in: ResizeImage

• AREA = 1

• BICUBIC = 2

• BILINEAR = 3

• NEAREST_NEIGHBOR = 4

Used in: PreprocessingStep

• optional int32 target_height = 1

• optional int32 target_width = 2

Used in: PreprocessingStep

• optional bool three_channels = 1

Used in: AttentionPredictor, BidirectionalRnn, LanguageModelRnnCell

• oneof rnn_cell_oneof

  • LstmCell lstm_cell = 1

  • GruCell gru_cell = 2

Used in: Loss

• optional bool sequence_normalize = 1

• optional bool sample_normalize = 2

• optional float weight = 3

Used in: MultiPredictorsRecognitionModel

• optional Convnet convnet = 1

• optional Hyperparams fc_hyperparams = 2

• optional int32 localization_h = 3

  image size for the localization network

• optional int32 localization_w = 4

• optional int32 output_h = 5

  rectified image size

• optional int32 output_w = 6

• optional float margin_x = 7

• optional float margin_y = 8

• optional int32 num_control_points = 9

• optional string init_bias_pattern = 10

• optional string activation = 11

• optional bool summarize_activations = 12

Used in: Convnet

• optional Hyperparams conv_hyperparams = 1

• optional bool summarize_activations = 2

• optional bool tiny = 3

Used in: PreprocessingStep

• optional bool lower_case = 1

• optional CharacterSet include_charset = 2

Used in: PreprocessingStep

• repeated float means = 1

  The mean to subtract from each channel. Should be of same dimension of channels in the input image.

An input reader that reads TF Example protos from local TFRecord files.

Used in: InputReader

• optional string input_path = 1

  Path to TFRecordFile.

Used in: MultiPredictorsRecognitionModel

• repeated BidirectionalRnn bidirectional_rnn = 1

• optional Convnet convnet = 2

• optional Loss loss = 3

• optional bool summarize_activations = 4

Used in: Convnet

• optional Hyperparams conv_hyperparams = 1

• optional bool summarize_activations = 2

Message for configuring DetectionModel training jobs (train.py).

Used in: TrainEvalPipelineConfig

• repeated uint32 batch_size = 1

  Input queue batch size.

• repeated PreprocessingStep data_augmentation_options = 2

  Data augmentation options.

• optional bool sync_replicas = 3

  Whether to synchronize replicas during training.

• optional uint32 keep_checkpoint_every_n_hours = 4

  How frequently to keep checkpoints.

• optional Optimizer optimizer = 5

  Optimizer used to train the DetectionModel.

• optional float gradient_clipping_by_norm = 6

  If greater than 0, clips gradients by this value.

• optional string fine_tune_checkpoint = 7

  Checkpoint to restore variables from. Typically used to load feature extractor variables trained outside of object detection.

• optional bool from_detection_checkpoint = 8

  Specifies if the finetune checkpoint is from an object detection model. If from an object detection model, the model being trained should have the same parameters with the exception of the num_classes parameter. If false, it assumes the checkpoint was a object classification model.

• optional uint32 num_steps = 9

  Number of steps to train the DetectionModel for. If 0, will train the model indefinitely.

• optional float startup_delay_steps = 10

  Number of training steps between replica startup. This flag must be set to 0 if sync_replicas is set to true.

• optional float bias_grad_multiplier = 11

  If greater than 0, multiplies the gradient of bias variables by this amount.

• repeated string freeze_variables = 12

  Variables that should not be updated during training.

• optional int32 replicas_to_aggregate = 13

  Number of replicas to aggregate before making parameter updates.

• optional int32 batch_queue_capacity = 14

  Maximum number of elements to store within a queue.

• optional int32 num_batch_queue_threads = 15

  Number of threads to use for batching.

• optional int32 prefetch_queue_capacity = 16

  Maximum capacity of the queue used to prefetch assembled batches.

• optional int32 save_checkpoint_secs = 17

  Save checkpoint every n seconds

• optional int32 save_summaries_steps = 18

  save summaries every n steps

Convenience message for configuring a training and eval pipeline. Allows all of the pipeline parameters to be configured from one file.

• optional Model model = 1

• optional TrainConfig train_config = 2

• repeated InputReader train_input_reader = 3

• optional EvalConfig eval_config = 4

• optional InputReader eval_input_reader = 5

Configuration proto for truncated normal initializer. See https://www.tensorflow.org/api_docs/python/tf/truncated_normal_initializer

Used in: Initializer

• optional float mean = 1

• optional float stddev = 2

Used in: Initializer

• optional float minval = 1

• optional float maxval = 2

Configuration proto for variance scaling initializer. See https://www.tensorflow.org/api_docs/python/tf/contrib/layers/ variance_scaling_initializer

Used in: Initializer

• optional float factor = 1

• optional bool uniform = 2

• optional VarianceScalingInitializer.Mode mode = 3

Used in: VarianceScalingInitializer

• FAN_IN = 0

• FAN_OUT = 1

• FAN_AVG = 2