easy_vision.python.core.encoders.nets

easy_vision.python.core.encoders.nets.base_text_conv_encoder

class easy_vision.python.core.encoders.nets.base_text_conv_encoder.BaseTextConvEncoder(config, is_training)

Bases: object

Text Base Encoder Network

__init__(config, is_training)

x.__init__(…) initializes x; see help(type(x)) for signature

build_graph(inputs, inputs_true_shape=None)

Build conv encoder graph :param inputs: encoder input feature :param inputs_true_shape: true shape of batched images. [batch_size, 3].

Returns:encoder output feature feature_dict: a dict of encoder feature Return type:outputs

easy_vision.python.core.encoders.nets.text_conv5_encoder

class easy_vision.python.core.encoders.nets.text_conv5_encoder.TextConv5Encoder(config, is_training)

Bases: easy_vision.python.core.encoders.nets.base_text_conv_encoder.BaseTextConvEncoder

__init__(config, is_training)

x.__init__(…) initializes x; see help(type(x)) for signature

build_graph(inputs, inputs_true_shape=None)

Build conv encoder graph :param inputs: encoder input feature :param inputs_true_shape: true shape of batched images. [batch_size, 3].

Returns:encoder output feature feature_dict: a dict of encoder feature Return type:outputs

easy_vision.python.core.encoders.nets.text_senet5_encoder

class easy_vision.python.core.encoders.nets.text_senet5_encoder.TextSENet5Encoder(config, is_training)

Bases: easy_vision.python.core.encoders.nets.base_text_conv_encoder.BaseTextConvEncoder

__init__(config, is_training)

x.__init__(…) initializes x; see help(type(x)) for signature

build_graph(inputs, inputs_true_shape=None)

Build conv encoder graph :param inputs: encoder input feature :param inputs_true_shape: true shape of batched images. [batch_size, 3].

Returns:encoder output feature feature_dict: a dict of encoder feature Return type:outputs

easy_vision.python.core.encoders.nets.text_vgg_bai_encoder

class easy_vision.python.core.encoders.nets.text_vgg_bai_encoder.TextVGGBaiEncoder(config, is_training)

Bases: easy_vision.python.core.encoders.nets.base_text_conv_encoder.BaseTextConvEncoder

TextVGGBaiNet

__init__(config, is_training)

x.__init__(…) initializes x; see help(type(x)) for signature

build_graph(inputs, inputs_true_shape=None)

Build conv encoder graph :param inputs: encoder input feature :param inputs_true_shape: true shape of batched images. [batch_size, 3].

Returns:encoder output feature feature_dict: a dict of encoder feature Return type:outputs