Default package

Get desktop application:
View/edit binary Protocol Buffers messages

Used in: Optimizer

• optional float initial_accumulator_value = 1

• optional bool use_locking = 2

Used in: Optimizer

• optional float beta1 = 1

• optional float beta2 = 2

• optional float epsilon = 3

• optional bool use_locking = 4

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.

• optional float midn_loss_weight = 91

  MIDN loss weight.

• optional float oicr_loss_weight = 92

  OICR loss weight.

• optional FRCNN frcnn_options = 1

  FRCNN configs.

• optional Hyperparams fc_hyperparams = 12

  Hyperparams of all the FC layers.

• optional int32 oicr_iterations = 21

  OICR iterations.

• optional float oicr_iou_threshold = 22

  OICR IoU threshold.

• optional PostProcess midn_post_processor = 31

  Config of the MIDN post processor.

• optional PostProcess oicr_post_processor = 32

  Config of the OICR post processor.

• repeated int32 eval_min_dimension = 34

  Resolutions at inference time.

• optional bool oicr_use_proba_r_given_c = 36

  If true, use proba_r_given_c as oicr_score_0.

• optional LabelExtractor label_extractor = 93

  Pseudo label extractor.

Used in: Reader

• repeated string input_pattern = 1

  Pattern of the input files.

• optional int32 interleave_cycle_length = 2

  Cycle length of interleave call.

• optional bool is_training = 3

  If True, repeat the dataset and shuffle the batch.

• optional int32 shuffle_buffer_size = 4

  Shuffle buffer size.

• optional int32 map_num_parallel_calls = 5

  Number of parallel calls.

• optional int32 prefetch_buffer_size = 6

  Prefetch buffer size.

• optional int32 batch_size = 7

  Batch size.

• optional bool decode_image = 11

  If false, do not decode image; For text classifier training.

• optional ImageResizer image_resizer = 12

  Image resizer.

• optional Preprocess preprocess_options = 13

  Preprocess options.

• optional int32 max_num_proposals = 14

  Pad to the max number of proposals to ensure static shape.

• repeated float batch_resize_scale_value = 15

  Randomly resize image according to these scales values.

• optional string shard_indicator = 16

  Shard to read, in the format of '0/3', '1/3', '2/3'. The denominator denotes the number of shards. The numerator denotes the shard to export.

Used in: ImageResizer

(message has no fields)

Used in: Pipeline

• optional int32 steps = 1

  Positive number of steps for which to evaluate model. if `None`, evaluate util `input_fn` raises an end-of-input exception.

• optional int32 start_delay_secs = 2

  Start evaluating after waiting for this many seconds.

• optional int32 throttle_secs = 3

  Do not re-evaluate unless the last evaluation was started at least this many seconds ago.

Used in: LabelExtractor

• optional string label_file = 1

Used in: LabelExtractor

• optional string label_file = 1

FRCNN configs.

Used in: Cap2DetModel

• optional FasterRcnnFeatureExtractor feature_extractor = 1

  Feature extractor config.

• optional bool inplace_batchnorm_update = 2

  Whether to update batch_norm inplace during training. This is required for batch norm to work correctly on TPUs. When this is false, user must add a control dependency on tf.GraphKeys.UPDATE_OPS for train/loss op in order to update the batch norm moving average parameters.

• optional int32 initial_crop_size = 3

  Output size (width and height are set to be the same) of the initial bilinear interpolation based cropping during ROI pooling.

• optional int32 maxpool_kernel_size = 4

  Kernel size of the max pool op on the cropped feature map during ROI pooling.

• optional int32 maxpool_stride = 5

  Stride of the max pool op on the cropped feature map during ROI pooling.

• optional float dropout_keep_prob = 6

  Keep probability of the dropout layer.

• optional bool dropout_on_feature_map = 7

  If true, add a dropout layer after extracting feature map.

• optional string checkpoint_path = 8

  Path to the pre-trained checkpoint.

Used in: FRCNN

• optional string type = 1

  Type of Faster R-CNN model (e.g., 'faster_rcnn_resnet101'; See builders/model_builder.py for expected types).

• optional int32 first_stage_features_stride = 2

  Output stride of extracted RPN feature map.

• optional bool batch_norm_trainable = 3

  Whether to update batch norm parameters during training or not. When training with a relative large batch size (e.g. 8), it could be desirable to enable batch norm update.

Used in: ImageResizer

• optional int32 height = 1

  Desired height of image in pixels.

• optional int32 width = 2

  Desired width of image in pixels.

Used in: Initializer

(message has no fields)

Used in: Initializer

(message has no fields)

Used in: Optimizer

• optional bool use_locking = 1

Used in: TrainConfig

• optional string scope = 1

  Variable scope that the multiplier is applied.

• optional float multiplier = 2

  A float number denoting the multiplier.

Used in: LabelExtractor, TextModel

• optional string label_file = 1

Configuration proto for the convolution op hyperparameters. See tensorflow/models/research/object_detection/protos/hyperparams.proto for details.

Used in: Cap2DetModel

• 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!

• optional bool regularize_depthwise = 6

  Whether depthwise convolutions should be regularized. If this parameter is NOT set then the conv hyperparams will default to the parent scope.

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: Cap2DetReader

• oneof image_resizer_oneof

  • DefaultResizer default_resizer = 1

  • FixedShapeResizer fixed_shape_resizer = 2

  • KeepAspectRatioResizer keep_aspect_ratio_resizer = 3

  • RandomScaleResizer random_scale_resizer = 4

Proto with one-of field for initializers.

Used in: Hyperparams

• oneof initializer_oneof

  • TruncatedNormalInitializer truncated_normal_initializer = 1

  • VarianceScalingInitializer variance_scaling_initializer = 2

  • RandomNormalInitializer random_normal_initializer = 3

  • GlorotNormalInitializer glorot_normal_initializer = 4

  • GlorotUniformInitializer glorot_uniform_initializer = 5

Used in: ImageResizer

• optional int32 min_dimension = 3

  Desired size of the smaller image dimension in pixels.

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

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: Cap2DetModel

• oneof label_extractor_oneof

  • GroundtruthExtractor groundtruth_extractor = 1

  • ExactMatchExtractor exact_match_extractor = 2

  • ExtendMatchExtractor extend_match_extractor = 3

  • WordVectorMatchExtractor word_vector_match_extractor = 4

  • TextClassifierMatchExtractor text_classifier_match_extractor = 5

Used in: TrainConfig

• optional int32 decay_steps = 1

  Decay the learning rate every `decay_steps`.

• optional float decay_rate = 2

  Decay rate.

• optional bool staircase = 3

  If true, decay the learning rate at discrete intervals.

Used in: Pipeline

(message has no fields)

Used in: Optimizer

• optional float momentum = 1

• optional bool use_locking = 2

• optional bool use_nesterov = 3

Used in: TrainConfig

• oneof optimizer

  • GradientDescentOptimizer sgd = 1

  • AdagradOptimizer adagrad = 2

  • AdamOptimizer adam = 3

  • RMSPropOptimizer rmsprop = 4

  • MomentumOptimizer momentum = 5

• optional Reader train_reader = 1

  Train reader.

• optional Reader eval_reader = 2

  Eval reader.

• optional Model model = 3

  Model config.

• optional string model_dir = 4

  Path to the model directory.

• optional TrainConfig train_config = 5

  Train config.

• optional EvalConfig eval_config = 6

  Eval config

Used in: Cap2DetModel

• optional float score_thresh = 1

  Score thresh hold for NMS.

• optional float iou_thresh = 2

  IoU threshold to check the overlap.

• optional int32 max_size_per_class = 3

  Maximum detections per class.

• optional int32 max_total_size = 4

  Maximum total detections.

Used in: Cap2DetReader

• optional float random_flip_left_right_prob = 1

  Random flip.

• optional float random_crop_prob = 2

  Random crop.

• optional float random_crop_min_scale = 3

• optional float random_brightness_prob = 4

  Random brightness.

• optional float random_brightness_max_delta = 5

• optional float random_contrast_prob = 6

  Random contrast.

• optional float random_contrast_lower = 7

• optional float random_contrast_upper = 8

• optional float random_hue_prob = 9

  Random hue.

• optional float random_hue_max_delta = 10

• optional float random_saturation_prob = 11

  Random saturation.

• optional float random_saturation_lower = 12

• optional float random_saturation_upper = 13

Used in: Optimizer

• optional float decay = 1

• optional float momentum = 2

• optional float epsilon = 3

• optional bool use_locking = 4

• optional bool centered = 5

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

Used in: Initializer

• optional float mean = 1

• optional float stddev = 2

Used in: ImageResizer

• repeated int32 max_dimension = 1

  Desired size of the larger image dimension in pixels.

Used in: Pipeline

• oneof reader_oneof

  • Cap2DetReader cap2det_reader = 1

Proto with one-of field for regularizers.

Used in: Hyperparams

• oneof regularizer_oneof

  • L1Regularizer l1_regularizer = 1

  • L2Regularizer l2_regularizer = 2

Used in: LabelExtractor, TextModel

• optional string label_file = 1

• optional string open_vocabulary_file = 2

  Path to the open-vocabulary file.

• optional string open_vocabulary_word_embedding_file = 3

  Path to the open-vocabulary word embedding file. This file should be loaded using `np.load`, resulting in a [vocab_size, embedding_dims] numpy array.

• optional string text_classifier_checkpoint_file = 4

  Path to the text classifier checkpoint file.

• optional int32 hidden_units = 5

  Number of hidden units.

• optional float dropout_keep_proba = 6

• optional float regularizer = 8

• optional float label_threshold = 7

  Minimum score to be considered as a label.

• optional GroundtruthExtractor label_extractor = 1

  Groundtruth extractor.

• optional TextClassifierMatchExtractor text_classifier = 2

  Text classifier.

Used in: Pipeline

• optional int32 max_steps = 1

  Positive number of total steps for which to train model.

• optional Optimizer optimizer = 2

  Optimizer to use.

• optional float learning_rate = 3

  Learning rate.

• optional int32 save_summary_steps = 4

  Save summaries every this many steps.

• optional int32 save_checkpoints_steps = 5

  Save checkpoints every this many steps.

• optional int32 keep_checkpoint_max = 6

  The maximum number of recent checkpoint files to keep.

• optional int32 log_step_count_steps = 7

  The frequency, in number of global steps, that the global step/sec and the loss will be logged during training.

• optional LearningRateDecay learning_rate_decay = 11

  Learning rate decay strategem.

• optional bool sync_replicas = 12

  If true, sync replicas using SyncReplicasOptimizer.

• optional float moving_average_decay = 13

  If set, enable to use moving average.

• repeated GradientMultiplier gradient_multiplier = 16

  Gradient multipliers.

• optional float max_gradient_norm = 17

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

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

Used in: LabelExtractor

• optional string label_file = 1

• optional string open_vocabulary_file = 2

  Path to the open-vocabulary file.

• optional string open_vocabulary_word_embedding_file = 3

  Path to the open-vocabulary word embedding file. This file should be loaded using `np.load`, resulting in a [vocab_size, embedding_dims] numpy array.