From be52f333ba226d6c4e44f798ae05e49807385326 Mon Sep 17 00:00:00 2001 From: DuYao Date: Fri, 11 Oct 2019 14:46:15 +0800 Subject: [PATCH] update English Documents, test=release/1.6, test=document_fix (#20482) --- paddle/fluid/API.spec | 24 +- .../fluid/dygraph/learning_rate_scheduler.py | 302 ++++++++++-------- python/paddle/fluid/dygraph/nn.py | 222 +++++++------ python/paddle/fluid/layers/nn.py | 32 +- 4 files changed, 336 insertions(+), 244 deletions(-) diff --git a/paddle/fluid/API.spec b/paddle/fluid/API.spec index 4a6629e603..7d28f27bd8 100644 --- a/paddle/fluid/API.spec +++ b/paddle/fluid/API.spec @@ -199,7 +199,7 @@ paddle.fluid.layers.lod_append (ArgSpec(args=['x', 'level'], varargs=None, keywo paddle.fluid.layers.lrn (ArgSpec(args=['input', 'n', 'k', 'alpha', 'beta', 'name'], varargs=None, keywords=None, defaults=(5, 1.0, 0.0001, 0.75, None)), ('document', 'fa565b65fb98d3ca82361c79f41b06b2')) paddle.fluid.layers.pad (ArgSpec(args=['x', 'paddings', 'pad_value', 'name'], varargs=None, keywords=None, defaults=(0.0, None)), ('document', '46b3ada86dd2c79042dca90a55e08f66')) paddle.fluid.layers.pad_constant_like (ArgSpec(args=['x', 'y', 'pad_value', 'name'], varargs=None, keywords=None, defaults=(0.0, None)), ('document', '89aa122a50dc20ee116ae49d66854d20')) -paddle.fluid.layers.label_smooth (ArgSpec(args=['label', 'prior_dist', 'epsilon', 'dtype', 'name'], varargs=None, keywords=None, defaults=(None, 0.1, 'float32', None)), ('document', '214f1dfbe95a628600bbe99e836319cf')) +paddle.fluid.layers.label_smooth (ArgSpec(args=['label', 'prior_dist', 'epsilon', 'dtype', 'name'], varargs=None, keywords=None, defaults=(None, 0.1, 'float32', None)), ('document', '70b6f4ab59e60650231b1ead4ad46222')) paddle.fluid.layers.roi_pool (ArgSpec(args=['input', 'rois', 'pooled_height', 'pooled_width', 'spatial_scale'], varargs=None, keywords=None, defaults=(1, 1, 1.0)), ('document', '6fc9bae94518bbf3e1a9e479f38f6537')) paddle.fluid.layers.roi_align (ArgSpec(args=['input', 'rois', 'pooled_height', 'pooled_width', 'spatial_scale', 'sampling_ratio', 'name'], varargs=None, keywords=None, defaults=(1, 1, 1.0, -1, None)), ('document', '3885fd76e122ac0563fa8369bcab7363')) paddle.fluid.layers.dice_loss (ArgSpec(args=['input', 'label', 'epsilon', 'name'], varargs=None, keywords=None, defaults=(1e-05, None)), ('document', '08d94daffbea3935178810bdc1633f07')) @@ -604,7 +604,7 @@ paddle.fluid.dygraph.Conv2D.set_dict (ArgSpec(args=['self', 'stat_dict', 'includ paddle.fluid.dygraph.Conv2D.state_dict (ArgSpec(args=['self', 'destination', 'include_sublayers'], varargs=None, keywords=None, defaults=(None, True)), ('document', '9d689f44592cd22812c7ec06a9654eac')) paddle.fluid.dygraph.Conv2D.sublayers (ArgSpec(args=['self', 'include_sublayers'], varargs=None, keywords=None, defaults=(True,)), ('document', '00a881005ecbc96578faf94513bf0d62')) paddle.fluid.dygraph.Conv2D.train (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.Conv3D ('paddle.fluid.dygraph.nn.Conv3D', ('document', '50412bd3fbf3557a8ef48e25c6517025')) +paddle.fluid.dygraph.Conv3D ('paddle.fluid.dygraph.nn.Conv3D', ('document', 'f81dee6781d6c18d0e7f5ca66b2fb010')) paddle.fluid.dygraph.Conv3D.__init__ (ArgSpec(args=['self', 'name_scope', 'num_filters', 'filter_size', 'stride', 'padding', 'dilation', 'groups', 'param_attr', 'bias_attr', 'use_cudnn', 'act'], varargs=None, keywords=None, defaults=(1, 0, 1, None, None, None, True, None)), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.Conv3D.add_parameter (ArgSpec(args=['self', 'name', 'parameter'], varargs=None, keywords=None, defaults=None), ('document', 'f35ab374c7d5165c3daf3bd64a5a2ec1')) paddle.fluid.dygraph.Conv3D.add_sublayer (ArgSpec(args=['self', 'name', 'sublayer'], varargs=None, keywords=None, defaults=None), ('document', '839ff3c0534677ba6ad8735c3fd4e995')) @@ -689,7 +689,7 @@ paddle.fluid.dygraph.Embedding.set_dict (ArgSpec(args=['self', 'stat_dict', 'inc paddle.fluid.dygraph.Embedding.state_dict (ArgSpec(args=['self', 'destination', 'include_sublayers'], varargs=None, keywords=None, defaults=(None, True)), ('document', '9d689f44592cd22812c7ec06a9654eac')) paddle.fluid.dygraph.Embedding.sublayers (ArgSpec(args=['self', 'include_sublayers'], varargs=None, keywords=None, defaults=(True,)), ('document', '00a881005ecbc96578faf94513bf0d62')) paddle.fluid.dygraph.Embedding.train (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.GRUUnit ('paddle.fluid.dygraph.nn.GRUUnit', ('document', '389e860e455b67aab1f4d472ac9d7e49')) +paddle.fluid.dygraph.GRUUnit ('paddle.fluid.dygraph.nn.GRUUnit', ('document', 'f0e648f0a8d3389f755698dde488dc93')) paddle.fluid.dygraph.GRUUnit.__init__ (ArgSpec(args=['self', 'name_scope', 'size', 'param_attr', 'bias_attr', 'activation', 'gate_activation', 'origin_mode', 'dtype'], varargs=None, keywords=None, defaults=(None, None, 'tanh', 'sigmoid', False, 'float32')), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.GRUUnit.add_parameter (ArgSpec(args=['self', 'name', 'parameter'], varargs=None, keywords=None, defaults=None), ('document', 'f35ab374c7d5165c3daf3bd64a5a2ec1')) paddle.fluid.dygraph.GRUUnit.add_sublayer (ArgSpec(args=['self', 'name', 'sublayer'], varargs=None, keywords=None, defaults=None), ('document', '839ff3c0534677ba6ad8735c3fd4e995')) @@ -757,7 +757,7 @@ paddle.fluid.dygraph.PRelu.set_dict (ArgSpec(args=['self', 'stat_dict', 'include paddle.fluid.dygraph.PRelu.state_dict (ArgSpec(args=['self', 'destination', 'include_sublayers'], varargs=None, keywords=None, defaults=(None, True)), ('document', '9d689f44592cd22812c7ec06a9654eac')) paddle.fluid.dygraph.PRelu.sublayers (ArgSpec(args=['self', 'include_sublayers'], varargs=None, keywords=None, defaults=(True,)), ('document', '00a881005ecbc96578faf94513bf0d62')) paddle.fluid.dygraph.PRelu.train (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.BilinearTensorProduct ('paddle.fluid.dygraph.nn.BilinearTensorProduct', ('document', 'be70d0f6d43729d9cb80c9a34ed5f26b')) +paddle.fluid.dygraph.BilinearTensorProduct ('paddle.fluid.dygraph.nn.BilinearTensorProduct', ('document', 'ddea5bc0668a636ded7db09538511c20')) paddle.fluid.dygraph.BilinearTensorProduct.__init__ (ArgSpec(args=['self', 'name_scope', 'size', 'name', 'act', 'param_attr', 'bias_attr'], varargs=None, keywords=None, defaults=(None, None, None, None)), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.BilinearTensorProduct.add_parameter (ArgSpec(args=['self', 'name', 'parameter'], varargs=None, keywords=None, defaults=None), ('document', 'f35ab374c7d5165c3daf3bd64a5a2ec1')) paddle.fluid.dygraph.BilinearTensorProduct.add_sublayer (ArgSpec(args=['self', 'name', 'sublayer'], varargs=None, keywords=None, defaults=None), ('document', '839ff3c0534677ba6ad8735c3fd4e995')) @@ -791,7 +791,7 @@ paddle.fluid.dygraph.Conv2DTranspose.set_dict (ArgSpec(args=['self', 'stat_dict' paddle.fluid.dygraph.Conv2DTranspose.state_dict (ArgSpec(args=['self', 'destination', 'include_sublayers'], varargs=None, keywords=None, defaults=(None, True)), ('document', '9d689f44592cd22812c7ec06a9654eac')) paddle.fluid.dygraph.Conv2DTranspose.sublayers (ArgSpec(args=['self', 'include_sublayers'], varargs=None, keywords=None, defaults=(True,)), ('document', '00a881005ecbc96578faf94513bf0d62')) paddle.fluid.dygraph.Conv2DTranspose.train (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.Conv3DTranspose ('paddle.fluid.dygraph.nn.Conv3DTranspose', ('document', '91ba132bc690eaf76eabdbde8f87e4a0')) +paddle.fluid.dygraph.Conv3DTranspose ('paddle.fluid.dygraph.nn.Conv3DTranspose', ('document', '0ef981fd6a74aaff21673f9925736ac7')) paddle.fluid.dygraph.Conv3DTranspose.__init__ (ArgSpec(args=['self', 'name_scope', 'num_filters', 'output_size', 'filter_size', 'padding', 'stride', 'dilation', 'groups', 'param_attr', 'bias_attr', 'use_cudnn', 'act', 'name'], varargs=None, keywords=None, defaults=(None, None, 0, 1, 1, None, None, None, True, None, None)), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.Conv3DTranspose.add_parameter (ArgSpec(args=['self', 'name', 'parameter'], varargs=None, keywords=None, defaults=None), ('document', 'f35ab374c7d5165c3daf3bd64a5a2ec1')) paddle.fluid.dygraph.Conv3DTranspose.add_sublayer (ArgSpec(args=['self', 'name', 'sublayer'], varargs=None, keywords=None, defaults=None), ('document', '839ff3c0534677ba6ad8735c3fd4e995')) @@ -870,31 +870,31 @@ paddle.fluid.dygraph.Tracer.train_mode (ArgSpec(args=['self'], varargs=None, key paddle.fluid.dygraph.prepare_context (ArgSpec(args=['strategy'], varargs=None, keywords=None, defaults=(None,)), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.save_dygraph (ArgSpec(args=['state_dict', 'model_path'], varargs=None, keywords=None, defaults=None), ('document', '7c2bd58a69f9bca3b884f44154c84569')) paddle.fluid.dygraph.load_dygraph (ArgSpec(args=['model_path'], varargs=None, keywords=None, defaults=None), ('document', 'd6d98002c39d2484835f4748e35b761c')) -paddle.fluid.dygraph.NoamDecay ('paddle.fluid.dygraph.learning_rate_scheduler.NoamDecay', ('document', '9ccfea97dbf15134d406a23aae1e1fa2')) +paddle.fluid.dygraph.NoamDecay ('paddle.fluid.dygraph.learning_rate_scheduler.NoamDecay', ('document', '3441619381487db8d1929a205f3c6d41')) paddle.fluid.dygraph.NoamDecay.__init__ (ArgSpec(args=['self', 'd_model', 'warmup_steps', 'begin', 'step', 'dtype'], varargs=None, keywords=None, defaults=(1, 1, 'float32')), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.NoamDecay.create_lr_var (ArgSpec(args=['self', 'lr'], varargs=None, keywords=None, defaults=None), ('document', '013bc233558149d0757b3df57845b866')) paddle.fluid.dygraph.NoamDecay.step (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.PiecewiseDecay ('paddle.fluid.dygraph.learning_rate_scheduler.PiecewiseDecay', ('document', '8f4d37eaad4e2f5b12850f3663856758')) +paddle.fluid.dygraph.PiecewiseDecay ('paddle.fluid.dygraph.learning_rate_scheduler.PiecewiseDecay', ('document', '0fccf303b94a13ae670fb3dd51931f73')) paddle.fluid.dygraph.PiecewiseDecay.__init__ (ArgSpec(args=['self', 'boundaries', 'values', 'begin', 'step', 'dtype'], varargs=None, keywords=None, defaults=(1, 'float32')), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.PiecewiseDecay.create_lr_var (ArgSpec(args=['self', 'lr'], varargs=None, keywords=None, defaults=None), ('document', '013bc233558149d0757b3df57845b866')) paddle.fluid.dygraph.PiecewiseDecay.step (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.NaturalExpDecay ('paddle.fluid.dygraph.learning_rate_scheduler.NaturalExpDecay', ('document', '94bed58b392a5a71b6d1abd39eed7111')) +paddle.fluid.dygraph.NaturalExpDecay ('paddle.fluid.dygraph.learning_rate_scheduler.NaturalExpDecay', ('document', '5fef27468d49ca8ca6c6a9635ad0f5c1')) paddle.fluid.dygraph.NaturalExpDecay.__init__ (ArgSpec(args=['self', 'learning_rate', 'decay_steps', 'decay_rate', 'staircase', 'begin', 'step', 'dtype'], varargs=None, keywords=None, defaults=(False, 0, 1, 'float32')), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.NaturalExpDecay.create_lr_var (ArgSpec(args=['self', 'lr'], varargs=None, keywords=None, defaults=None), ('document', '013bc233558149d0757b3df57845b866')) paddle.fluid.dygraph.NaturalExpDecay.step (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.ExponentialDecay ('paddle.fluid.dygraph.learning_rate_scheduler.ExponentialDecay', ('document', 'a259689c649c5f82636536386ce2ef19')) +paddle.fluid.dygraph.ExponentialDecay ('paddle.fluid.dygraph.learning_rate_scheduler.ExponentialDecay', ('document', '846eb564df136d8a8917bf16b5b8ac9b')) paddle.fluid.dygraph.ExponentialDecay.__init__ (ArgSpec(args=['self', 'learning_rate', 'decay_steps', 'decay_rate', 'staircase', 'begin', 'step', 'dtype'], varargs=None, keywords=None, defaults=(False, 0, 1, 'float32')), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.ExponentialDecay.create_lr_var (ArgSpec(args=['self', 'lr'], varargs=None, keywords=None, defaults=None), ('document', '013bc233558149d0757b3df57845b866')) paddle.fluid.dygraph.ExponentialDecay.step (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.InverseTimeDecay ('paddle.fluid.dygraph.learning_rate_scheduler.InverseTimeDecay', ('document', '6a868b2c7cc0f09f57ef71902bbc93ca')) +paddle.fluid.dygraph.InverseTimeDecay ('paddle.fluid.dygraph.learning_rate_scheduler.InverseTimeDecay', ('document', '1a74f0370e2e64f9e786d3c336526e6d')) paddle.fluid.dygraph.InverseTimeDecay.__init__ (ArgSpec(args=['self', 'learning_rate', 'decay_steps', 'decay_rate', 'staircase', 'begin', 'step', 'dtype'], varargs=None, keywords=None, defaults=(False, 0, 1, 'float32')), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.InverseTimeDecay.create_lr_var (ArgSpec(args=['self', 'lr'], varargs=None, keywords=None, defaults=None), ('document', '013bc233558149d0757b3df57845b866')) paddle.fluid.dygraph.InverseTimeDecay.step (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.PolynomialDecay ('paddle.fluid.dygraph.learning_rate_scheduler.PolynomialDecay', ('document', 'bb90314cee58952f13522dcd571ca832')) +paddle.fluid.dygraph.PolynomialDecay ('paddle.fluid.dygraph.learning_rate_scheduler.PolynomialDecay', ('document', 'e222a066a2bcf31bc52a14271048e034')) paddle.fluid.dygraph.PolynomialDecay.__init__ (ArgSpec(args=['self', 'learning_rate', 'decay_steps', 'end_learning_rate', 'power', 'cycle', 'begin', 'step', 'dtype'], varargs=None, keywords=None, defaults=(0.0001, 1.0, False, 0, 1, 'float32')), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.PolynomialDecay.create_lr_var (ArgSpec(args=['self', 'lr'], varargs=None, keywords=None, defaults=None), ('document', '013bc233558149d0757b3df57845b866')) paddle.fluid.dygraph.PolynomialDecay.step (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) -paddle.fluid.dygraph.CosineDecay ('paddle.fluid.dygraph.learning_rate_scheduler.CosineDecay', ('document', '46dadadee1a8a92d70bd277d9345bfb0')) +paddle.fluid.dygraph.CosineDecay ('paddle.fluid.dygraph.learning_rate_scheduler.CosineDecay', ('document', '0d7fe2b87492a0eb5cde60dbe268ea17')) paddle.fluid.dygraph.CosineDecay.__init__ (ArgSpec(args=['self', 'learning_rate', 'step_each_epoch', 'epochs', 'begin', 'step', 'dtype'], varargs=None, keywords=None, defaults=(0, 1, 'float32')), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) paddle.fluid.dygraph.CosineDecay.create_lr_var (ArgSpec(args=['self', 'lr'], varargs=None, keywords=None, defaults=None), ('document', '013bc233558149d0757b3df57845b866')) paddle.fluid.dygraph.CosineDecay.step (ArgSpec(args=['self'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754')) diff --git a/python/paddle/fluid/dygraph/learning_rate_scheduler.py b/python/paddle/fluid/dygraph/learning_rate_scheduler.py index 57d602afdb..f54654f8ad 100644 --- a/python/paddle/fluid/dygraph/learning_rate_scheduler.py +++ b/python/paddle/fluid/dygraph/learning_rate_scheduler.py @@ -69,30 +69,33 @@ class LearningRateDecay(object): class PiecewiseDecay(LearningRateDecay): """ - piecewise decay scheduler + Piecewise decay scheduler. The algorithm can be described as the code below. .. code-block:: text - boundaries = [10000, 20000] - values = [1.0, 0.5, 0.1] - if step < 10000: - learning_rate = 1.0 - elif 10000 <= step < 20000: - learning_rate = 0.5 - else: - learning_rate = 0.1 - Args: - boundaries: A list of steps numbers. - values: A list of learning rate values that will be picked during - different step boundaries. - begin: The begin step to initilize the self.step_num - step: The step_size using when calculate the new step_num (Defalult is 1) - dtype: The dtype used to create the learning rate variable + boundaries = [10000, 20000] + values = [1.0, 0.5, 0.1] + if global_step < 10000: + learning_rate = 1.0 + elif 10000 <= global_step < 20000: + learning_rate = 0.5 + else: + learning_rate = 0.1 + + Parameters: + boundaries(list): A list of steps numbers. The type of element in the list is python int. + values(list): A list of learning rate values that will be picked during + different step boundaries. The type of element in the list is python float. + begin(int): The begin step to initilize the global_step in the description above. + step(int, optional): The step size used to calculate the new global_step in the description above. + The defalult value is 1. + dtype(str, optional): The data type used to create the learning rate variable. The data type can be set as + 'float32', 'float64'. The default value is 'float32'. Returns: - The decayed learning rate. + None. Examples: .. code-block:: python @@ -125,25 +128,40 @@ class NaturalExpDecay(LearningRateDecay): """ Applies natural exponential decay to the initial learning rate. - .. code-block:: python + The algorithm can be described as following. - if not staircase: - decayed_learning_rate = learning_rate * exp(- decay_rate * (global_step / decay_steps)) - else: - decayed_learning_rate = learning_rate * exp(- decay_rate * (global_step / decay_steps)) - - Args: - learning_rate: A scalar float32 value or a Variable. This - will be the initial learning rate during training - decay_steps: A Python `int32` number. - decay_rate: A Python `float` number. - staircase: Boolean. If set true, decay the learning rate every decay_steps. - begin: A Python 'int32' number, the begin step (Default is 0) - step: A Python 'int32' number, the step size (Default is 1) - dtype: A Python 'str', the dtype used to create learning rate variable (Default is 'float32') + .. math:: + + decayed\_learning\_rate = learning\_rate * e^{y} + + If staircase is set to False, then: + + .. math:: + + y = - decay\_rate * \\frac{global\_step}{decay\_steps} + + If staircase is set to True, then: + + .. math:: + + y = - decay\_rate * math.floor(\\frac{global\_step}{decay\_steps}) + + Parameters: + learning_rate(Variable|float): The initial learning rate. If the type + is Variable, it's a tensor with shape [1], the data type can be + float32 or float64. It also can be set to python int number. + decay_steps(int): The decay step size. It determines the decay cycle. + decay_rate(int): The decay rate. + staircase(bool, optional): If set to True, decay the learning rate at discrete intervals. The + default value is False. + begin(int, optional): The begin step. The initial value of global_step described above. The default value is 0. + step(int, optional): The step size used to calculate the new global_step in the description above. + The defalult value is 1. + dtype(str, optional): The data type used to create the learning rate variable. The data type can be set as + 'float32', 'float64'. The default value is 'float32'. Returns: - The decayed learning rate. + None. Examples: .. code-block:: python @@ -189,29 +207,41 @@ class ExponentialDecay(LearningRateDecay): """ Applies exponential decay to the learning rate. - When training a model, it is often recommended to lower the learning rate as the - training progresses. By using this function, the learning rate will be decayed by - 'decay_rate' every 'decay_steps' steps. + The algorithm can be described as following. - .. code-block:: python + .. math:: - if staircase == True: - decayed_learning_rate = learning_rate * decay_rate ^ floor(global_step / decay_steps) - else: - decayed_learning_rate = learning_rate * decay_rate ^ (global_step / decay_steps) - - Args: - learning_rate(Variable|float): The initial learning rate. - decay_steps(int): See the decay computation above. - decay_rate(float): The decay rate. See the decay computation above. - staircase(Boolean): If True, decay the learning rate at discrete intervals. - Default: False - begin(int): The begin step (default is 0) - step(int): The step size (default is 1) - dtype(str): The dtype used to create learning rate (default is 'float32') + decayed\_learning\_rate = learning\_rate * decay\_rate ^ y + + If staircase is set to False, then: + + .. math:: + + y = \\frac{global\_step}{decay\_steps} + + If staircase is set to True, then: + + .. math:: + + y = math.floor(\\frac{global\_step}{decay\_steps}) + + + Parameters: + learning_rate(Variable|float): The initial learning rate. If the type + is Variable, it's a tensor with shape [1], the data type can be + float32 or float64. It also can be set to python int number. + decay_steps(int): The decay step size. It determines the decay cycle. + decay_rate(float): The decay rate. + staircase(bool, optional): If set to True, decay the learning rate at discrete intervals. The + default value is False. + begin(int, optional): The begin step. The initial value of global_step described above. The default value is 0. + step(int, optional): The step size used to calculate the new global_step in the description above. + The defalult value is 1. + dtype(str, optional): The data type used to create the learning rate variable. The data type can be set as + 'float32', 'float64'. The default value is 'float32'. Returns: - The decayed learning rate. + None. Examples: .. code-block:: python @@ -257,27 +287,35 @@ class InverseTimeDecay(LearningRateDecay): """ Applies inverse time decay to the initial learning rate. - When training a model, it is often recommended to lower the learning rate as the - training progresses. By using this function, an inverse decay function will be - applied to the initial learning rate. - - >>> if staircase == True: - >>> decayed_learning_rate = learning_rate / (1 + decay_rate * floor(global_step / decay_step)) - >>> else: - >>> decayed_learning_rate = learning_rate / (1 + decay_rate * global_step / decay_step) - - Args: - learning_rate(Variable|float): The initial learning rate. - decay_steps(int): See the decay computation above. - decay_rate(float): The decay rate. See the decay computation above. - staircase(Boolean): If True, decay the learning rate at discrete intervals. - Default: False - begin(int): The begin step (default is 0) - step(int): The step size (default is 1) - dtype(str): The dtype used to create learning rate (default is 'float32') + The algorithm can be described as following. + If staircase is set to False, then: + + .. math:: + + decayed\_learning\_rate = \\frac{learning\_rate}{1 + decay\_rate * \\frac{global\_step}{decay\_step}} + + If staircase is set to True, then: + + .. math:: + + decayed\_learning\_rate = \\frac{learning\_rate}{1 + decay\_rate * math.floor(\\frac{global\_step}{decay\_step})} + + Parameters: + learning_rate(Variable|float): The initial learning rate. If the type + is Variable, it's a tensor with shape [1], the data type can be + float32 or float64. It also can be set to python int number. + decay_steps(int): The decay step size. It determines the decay cycle. + decay_rate(float): The decay rate. + staircase(bool, optional): If set to True, decay the learning rate at discrete intervals. The + default value is False. + begin(int, optional): The begin step. The initial value of global_step described above. The default value is 0. + step(int, optional): The step size used to calculate the new global_step in the description above. + The defalult value is 1. + dtype(str, optional): The data type used to create the learning rate variable. The data type can be + 'float32', 'float64'. The default value is 'float32'. Returns: - The decayed learning rate. + None. Examples: .. code-block:: python @@ -323,28 +361,40 @@ class PolynomialDecay(LearningRateDecay): """ Applies polynomial decay to the initial learning rate. - .. code-block:: text + The algorithm can be described as following. + + If cycle is set to True, then: + + .. math:: + + decay\_steps & = decay\_steps * math.ceil(\\frac{global\_step}{decay\_steps}) - if cycle: - decay_steps = decay_steps * ceil(global_step / decay_steps) - else: - global_step = min(global_step, decay_steps) - decayed_learning_rate = (learning_rate - end_learning_rate) * - (1 - global_step / decay_steps) ^ power + end_learning_rate - - Args: - learning_rate(Variable|float32): A scalar float32 value or a Variable. This - will be the initial learning rate during training. - decay_steps(int32): A Python `int32` number. - end_learning_rate(float): A Python `float` number. - power(float): A Python `float` number. - cycle(bool): If set true, decay the learning rate every decay_steps. - begin(int): The begin step (default is 0) - step(int): The step size (default is 1) - dtype(str): The dtype used to create learning rate (default is 'float32') + decayed\_learning\_rate & = (learning\_rate-end\_learning\_rate)*(1-\\frac{global\_step}{decay\_steps})^{power}+end\_learning\_rate + + If cycle is set to False, then: + + .. math:: + + global\_step & = min(global\_step, decay\_steps) + + decayed\_learning\_rate & = (learning\_rate-end\_learning\_rate)*(1-\\frac{global\_step}{decay\_steps})^{power}+end\_learning\_rate + + Parameters: + learning_rate(Variable|float): The initial learning rate. If the type + is Variable, it's a tensor with shape [1], the data type can be + float32 or float64. It also can be set to python int number. + decay_steps(int32): The decay step size. It determines the decay cycle. + end_learning_rate(float, optional): The minimum final learning rate. The default value is 0.0001. + power(float, optional): Power of polynomial. The default value is 1.0. + cycle(bool, optional): If set true, decay the learning rate every decay_steps. The default value is False. + begin(int, optional): The begin step. The initial value of global_step described above. The default value is 0. + step(int, optional): The step size used to calculate the new global_step in the description above. + The defalult value is 1. + dtype(str, optional): The data type used to create the learning rate variable. The data type can be set as + 'float32', 'float64'. The default value is 'float32'. Returns: - The decayed learning rate. + None. Examples: .. code-block:: python @@ -401,24 +451,26 @@ class CosineDecay(LearningRateDecay): """ Applies cosine decay to the learning rate. - when training a model, it is often recommended to lower the learning rate as the - training progresses. By using this function, the learning rate will be decayed by - following cosine decay strategy. + The algorithm can be described as following. .. math:: - decayed\_lr = learning\_rate * 0.5 * (math.cos * (epoch * \\frac{math.pi}{epochs} ) + 1) + decayed\_learning\_rate = learning\_rate * 0.5 * (math.cos(global\_step * \\frac{math.pi}{step\_each\_epoch} ) + 1) - Args: - learning_rate(Variable|float): The initial learning rate. - step_each_epoch(int): the number of steps in an epoch. - epochs(int): the number of epochs. - begin(int): The begin step (default is 0). - step(int): The step size (default is 1). - dtype(str): The dtype used to create learning rate (default is 'float32'). + Parameters: + learning_rate(Variable|float): The initial learning rate. If the type + is Variable, it's a tensor with shape [1], the data type can be + float32 or float64. It also can be set to python int number. + step_each_epoch(int): The number of steps in an epoch. + epochs(int): The number of epochs. + begin(int, optional): The begin step. The initial value of global_step described above. The default value is 0. + step(int, optional): The step size used to calculate the new global_step in the description above. + The defalult value is 1. + dtype(str, optional): The data type used to create the learning rate variable. The data type can be set as + 'float32', 'float64'. The default value is 'float32'. Returns: - The decayed learning rate. + None. Examples: .. code-block:: python @@ -453,33 +505,29 @@ class CosineDecay(LearningRateDecay): class NoamDecay(LearningRateDecay): """ - Noam decay method. The numpy implementation of noam decay as follows. - - .. code-block:: python - - import numpy as np - # set hyper parameters - d_model = 2 - current_steps = 20 - warmup_steps = 200 - # compute - lr_value = np.power(d_model, -0.5) * np.min([ - np.power(current_steps, -0.5), - np.power(warmup_steps, -1.5) * current_steps]) - - Please reference `attention is all you need - `_. - - Args: - d_model(Variable): The dimensionality of input and output of model. - - warmup_steps(Variable): A super parameter. - begin(int): The begin step (default is 0) - step(int): The step size (default is 1) - dtype(str): The dtype used to create learning rate (default is 'float32') + Applies Noam decay to the initial learning rate. + + The algorithm can be described as following. + + .. math:: + + decayed\_learning\_rate = d_{model}^{-0.5} * min(global\_step^{-0.5}, global\_step * warmup\_steps^{-1.5}) + + Please reference `attention is all you need `_ + + Parameters: + d$_{model}$(Variable|int): The dimensionality of input and output feature vector of model. If type is Variable, + it's a tensor with shape [1] and the data type can be int32 or int64. The type can also be python int. + warmup_steps(Variable|int): The number of warmup steps. A super parameter. If type is Variable, + it's a tensor with shape [1] and the data type can be int32 or int64. The type can also be python int. + begin(int, optional): The begin step. The initial value of global_step described above. The default value is 0. + step(int, optional): The step size used to calculate the new global_step in the description above. + The defalult value is 1. + dtype(str, optional): The data type used to create the learning rate variable. The data type can be set as + 'float32', 'float64'. The default value is 'float32'. Returns: - The decayed learning rate. + None. Examples: .. code-block:: python diff --git a/python/paddle/fluid/dygraph/nn.py b/python/paddle/fluid/dygraph/nn.py index 6cce4727e2..eb76463292 100644 --- a/python/paddle/fluid/dygraph/nn.py +++ b/python/paddle/fluid/dygraph/nn.py @@ -273,7 +273,8 @@ class Conv3D(layers.Layer): The convolution3D layer calculates the output based on the input, filter and strides, paddings, dilations, groups parameters. Input(Input) and - Output(Output) are in NCDHW format. Where N is batch size C is the number of + Output(Output) are multidimensional tensors with a shape of + :math:`[N, C, D, H, W]` . Where N is batch size, C is the number of channels, D is the depth of the feature, H is the height of the feature, and W is the width of the feature. Convlution3D is similar with Convlution2D but adds one dimension(depth). If bias attribution and activation type are @@ -288,7 +289,7 @@ class Conv3D(layers.Layer): In the above equation: - * :math:`X`: Input value, a tensor with NCDHW format. + * :math:`X`: Input value, a tensor with NCDHW or NDHWC format. * :math:`W`: Filter value, a tensor with MCDHW format. * :math:`\\ast`: Convolution operation. * :math:`b`: Bias value, a 2-D tensor with shape [M, 1]. @@ -317,45 +318,46 @@ class Conv3D(layers.Layer): Parameters: name_scope(str) : The name for this class. num_filters(int): The number of filter. It is as same as the output image channel. - filter_size (int|tuple|None): The filter size. If filter_size is a tuple, + filter_size (int|tuple, optional): The filter size. If filter_size is a tuple, it must contain three integers, (filter_size_D, filter_size_H, filter_size_W). - Otherwise, the filter will be a square. - stride (int|tuple): The stride size. If stride is a tuple, it must + Otherwise, the filter will be a square, filter_size_depth = filter_size_height + = filter_size_width = filter_size. + stride (int|tuple, optional): The stride size. If stride is a tuple, it must contain three integers, (stride_D, stride_H, stride_W). Otherwise, the - stride_D = stride_H = stride_W = stride. Default: stride = 1. - padding (int|tuple): The padding size. If padding is a tuple, it must + stride_D = stride_H = stride_W = stride. The default value is 1. + padding (int|tuple, optional): The padding size. If padding is a tuple, it must contain three integers, (padding_D, padding_H, padding_W). Otherwise, the - padding_D = padding_H = padding_W = padding. Default: padding = 0. - dilation (int|tuple): The dilation size. If dilation is a tuple, it must + padding_D = padding_H = padding_W = padding. The default value is 0. + dilation (int|tuple, optional): The dilation size. If dilation is a tuple, it must contain three integers, (dilation_D, dilation_H, dilation_W). Otherwise, the - dilation_D = dilation_H = dilation_W = dilation. Default: dilation = 1. - groups (int): The groups number of the Conv3d Layer. According to grouped + dilation_D = dilation_H = dilation_W = dilation. The default value is 1. + groups (int, optional): The groups number of the Conv3d Layer. According to grouped convolution in Alex Krizhevsky's Deep CNN paper: when group=2, the first half of the filters is only connected to the first half of the input channels, while the second half of the filters is only - connected to the second half of the input channels. Default: groups=1 - param_attr (ParamAttr|None): The parameter attribute for learnable parameters/weights + connected to the second half of the input channels. The default value is 1. + param_attr (ParamAttr, optional): The parameter attribute for learnable parameters/weights of conv3d. If it is set to None or one attribute of ParamAttr, conv3d will create ParamAttr as param_attr. If it is set to None, the parameter is initialized with :math:`Normal(0.0, std)`, and the :math:`std` is - :math:`(\\frac{2.0 }{filter\_elem\_num})^{0.5}`. Default: None. - bias_attr (ParamAttr|bool|None): The parameter attribute for the bias of conv3d. + :math:`(\\frac{2.0 }{filter\_elem\_num})^{0.5}`. The default value is None. + bias_attr (ParamAttr|bool, optional): The parameter attribute for the bias of conv3d. If it is set to False, no bias will be added to the output units. If it is set to None or one attribute of ParamAttr, conv3d will create ParamAttr as bias_attr. If the Initializer of the bias_attr - is not set, the bias is initialized zero. Default: None. - use_cudnn (bool): Use cudnn kernel or not, it is valid only when the cudnn - library is installed. Default: True - act (str): Activation type, if it is set to None, activation is not appended. - Default: None. + is not set, the bias is initialized zero. The default value is None. + use_cudnn (bool, optional): Use cudnn kernel or not, it is valid only when the cudnn + library is installed. The default value is True. + act (str, optional): Activation type, if it is set to None, activation is not appended. + The default value is None. - Attributes: - weight (Parameter): the learnable weights of filters of this layer. - bias (Parameter|None): the learnable bias of this layer. + Attribute: + **weight** (Parameter): the learnable weights of filters of this layer. + + **bias** (Parameter): the learnable bias of this layer. Returns: - Variable: The tensor variable storing the convolution and \ - non-linearity activation result. + None. Raises: ValueError: If the shapes of input, filter_size, stride, padding and @@ -533,58 +535,85 @@ class Conv3DTranspose(layers.Layer): .. math:: - D_{out} &= (D_{in} - 1) * strides[0] - 2 * paddings[0] + dilations[0] * (D_f - 1) + 1 \\\\ - H_{out} &= (H_{in} - 1) * strides[1] - 2 * paddings[1] + dilations[1] * (H_f - 1) + 1 \\\\ - W_{out} &= (W_{in} - 1) * strides[2] - 2 * paddings[2] + dilations[2] * (W_f - 1) + 1 + D^\prime_{out} &= (D_{in} - 1) * strides[0] - 2 * paddings[0] + dilations[0] * (D_f - 1) + 1 \\\\ + H^\prime_{out} &= (H_{in} - 1) * strides[1] - 2 * paddings[1] + dilations[1] * (H_f - 1) + 1 \\\\ + W^\prime_{out} &= (W_{in} - 1) * strides[2] - 2 * paddings[2] + dilations[2] * (W_f - 1) + 1 \\\\ + D_{out} &\in [ D^\prime_{out}, D^\prime_{out} + strides[0] ] \\\\ + H_{out} &\in [ H^\prime_{out}, H^\prime_{out} + strides[1] ] \\\\ + + **Note**: + + The conv3d_transpose can be seen as the backward of the conv3d. For conv3d, + when stride > 1, conv3d maps multiple input shape to the same output shape, + so for conv3d_transpose, when stride > 1, input shape maps multiple output shape. + If output_size is None, :math:`H_{out} = H^\prime_{out}, :math:`H_{out} = \ + H^\prime_{out}, W_{out} = W^\prime_{out}`; else, the :math:`D_{out}` of the output + size must between :math:`D^\prime_{out}` and :math:`D^\prime_{out} + strides[0]`, + the :math:`H_{out}` of the output size must between :math:`H^\prime_{out}` + and :math:`H^\prime_{out} + strides[1]`, and the :math:`W_{out}` of the output size must + between :math:`W^\prime_{out}` and :math:`W^\prime_{out} + strides[2]`, + conv3d_transpose can compute the kernel size automatically. + Parameters: name_scope(str) : The name for this class. num_filters(int): The number of the filter. It is as same as the output image channel. - output_size(int|tuple|None): The output image size. If output size is a - tuple, it must contain three integers, (image_D, image_H, image_W). This - parameter only works when filter_size is None. - filter_size(int|tuple|None): The filter size. If filter_size is a tuple, + output_size(int|tuple, optional): The output image size. If output size is a + tuple, it must contain three integers, (image_depth, image_height, image_width). This + parameter only works when filter_size is None. If output_size and filter_size are + specified at the same time, They should follow the formula above. The default value is None. + Output_size and filter_size should not be None at the same time. + filter_size(int|tuple, optional): The filter size. If filter_size is a tuple, it must contain three integers, (filter_size_D, filter_size_H, filter_size_W). Otherwise, the filter will be a square. None if use output size to - calculate filter_size. - padding(int|tuple): The padding size. If padding is a tuple, it must - contain three integers, (padding_D, padding_H, padding_W). Otherwise, the - padding_D = padding_H = padding_W = padding. Default: padding = 0. - stride(int|tuple): The stride size. If stride is a tuple, it must - contain three integers, (stride_D, stride_H, stride_W). Otherwise, the - stride_D = stride_H = stride_W = stride. Default: stride = 1. - dilation(int|tuple): The dilation size. If dilation is a tuple, it must + calculate filter_size. The default value is None. + padding(int|tuple, optional): The padding size. The padding argument effectively + adds `dilation * (kernel - 1)` amount of zero-padding on both sides of input. If `padding` is a string, + either 'VALID' or 'SAME' supported, which is the padding algorithm. If `padding` + is a tuple or list, it could be in three forms: `[pad_depth, pad_height, pad_width]` or + `[pad_depth_front, pad_depth_back, pad_height_top, pad_height_bottom, pad_width_left, pad_width_right]`, + and when `data_format` is `'NCDHW'`, `padding` can be in the form + `[[0,0], [0,0], [pad_depth_front, pad_depth_back], [pad_height_top, pad_height_bottom], [pad_width_left, pad_width_right]]`. + when `data_format` is `'NDHWC'`, `padding` can be in the form + `[[0,0], [pad_depth_front, pad_depth_back], [pad_height_top, pad_height_bottom], [pad_width_left, pad_width_right], [0,0]]`. + The default value is 0. + stride(int|tuple, optional): The stride size. It means the stride in transposed convolution. + If stride is a tuple, it must contain three integers, (stride_depth, stride_height, + stride_width). Otherwise, stride_depth = stride_height = stride_width = stride. + The default value is 1. + dilation(int|tuple, optional): The dilation size. If dilation is a tuple, it must contain three integers, (dilation_D, dilation_H, dilation_W). Otherwise, the - dilation_D = dilation_H = dilation_W = dilation. Default: dilation = 1. - groups(int): The groups number of the Conv3d transpose layer. Inspired by + dilation_D = dilation_H = dilation_W = dilation. The default value is 1. + groups(int, optional): The groups number of the Conv3d transpose layer. Inspired by grouped convolution in Alex Krizhevsky's Deep CNN paper, in which when group=2, the first half of the filters is only connected to the first half of the input channels, while the second half of the filters is only connected to the second half of the input channels. - Default: groups=1 - param_attr (ParamAttr|None): The parameter attribute for learnable parameters/weights + The default value is 1. + param_attr (ParamAttr, optional): The parameter attribute for learnable parameters/weights of conv3d_transpose. If it is set to None or one attribute of ParamAttr, conv3d_transpose will create ParamAttr as param_attr. If the Initializer of the param_attr - is not set, the parameter is initialized with Xavier. Default: None. - bias_attr (ParamAttr|bool|None): The parameter attribute for the bias of conv3d_transpose. + is not set, the parameter is initialized with Xavier. The default value is None. + bias_attr (ParamAttr|bool, optional): The parameter attribute for the bias of conv3d_transpose. If it is set to False, no bias will be added to the output units. If it is set to None or one attribute of ParamAttr, conv3d_transpose will create ParamAttr as bias_attr. If the Initializer of the bias_attr - is not set, the bias is initialized zero. Default: None. - use_cudnn(bool): Use cudnn kernel or not, it is valid only when the cudnn - library is installed. Default: True - act (str): Activation type, if it is set to None, activation is not appended. - Default: None. - name(str|None): A name for this layer(optional). If set None, the layer - will be named automatically. + is not set, the bias is initialized zero. The default value is None. + use_cudnn(bool, optional): Use cudnn kernel or not, it is valid only when the cudnn + library is installed. The default value is True. + act (str, optional): Activation type, if it is set to None, activation is not appended. + The default value is None. + name(str, optional): The default value is None. Normally there is no need for user + to set this property. For more information, please refer to :ref:`api_guide_Name`. - Attributes: - weight (Parameter): the learnable weights of filters of this layer. - bias (Parameter|None): the learnable bias of this layer. + Attribute: + **weight** (Parameter): the learnable weights of filters of this layer. + + **bias** (Parameter): the learnable bias of this layer. Returns: - Variable: The tensor variable storing the convolution transpose result. + None. Raises: ValueError: If the shapes of input, filter_size, stride, padding and @@ -1626,10 +1655,11 @@ class LayerNorm(layers.Layer): class GRUUnit(layers.Layer): """ **GRU unit layer** - - if origin_mode is True, then the equation of a gru step is from paper - `Learning Phrase Representations using RNN Encoder-Decoder for Statistical - Machine Translation ` + + It creates a callable object from GRUUnit class. + If origin_mode is True, then the equation of a gru step is from paper + `Learning Phrase Representations using RNN Encoder-Decoder for Statistical + Machine Translation `_ .. math:: u_t & = actGate(xu_{t} + W_u h_{t-1} + b_u) @@ -1640,7 +1670,7 @@ class GRUUnit(layers.Layer): h_t & = dot(u_t, h_{t-1}) + dot((1-u_t), m_t) - if origin_mode is False, then the equation of a gru step is from paper + If origin_mode is False, then the equation of a gru step is from paper `Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling `_ @@ -1669,39 +1699,46 @@ class GRUUnit(layers.Layer): Parameters: name_scope(str): The name of this class. size (int): The input dimension value. - param_attr(ParamAttr|None): The parameter attribute for the learnable - hidden-hidden weight matrix. Note: + param_attr(ParamAttr, optional): The parameter attribute for the learnable + hidden-hidden weight matrix. + + **Note**: + + 1. The shape of the weight matrix is :math:`[T, 3*D]`, where D is the hidden size. + 2. All elements in the weight matrix can be divided into two parts. The first + part are weights of the update gate and reset gate with shape :math:`[D, 2*D]`, + and the second part are weights for candidate hidden state with shape :math:`[D, D]`. - - The shape of the weight matrix is :math:`(T \\times 3D)`, where - :math:`D` is the hidden size. - - All elements in the weight matrix can be divided into two parts. - The first part are weights of the update gate and reset gate with - shape :math:`(D \\times 2D)`, and the second part are weights for - candidate hidden state with shape :math:`(D \\times D)`. If it is set to None or one attribute of ParamAttr, gru_unit will create ParamAttr as param_attr. If the Initializer of the param_attr - is not set, the parameter is initialized with Xavier. Default: None. - bias_attr (ParamAttr|bool|None): The parameter attribute for the bias - of GRU.Note that the bias with :math:`(1 \\times 3D)` concatenates + is not set, the parameter is initialized with Xavier. The default + value is None. + bias_attr (ParamAttr|bool, optional): The parameter attribute for the bias + of GRU.Note that the bias with :math:`[1, 3*D]` concatenates the bias in the update gate, reset gate and candidate calculations. If it is set to False, no bias will be applied to the update gate, reset gate and candidate calculations. If it is set to None or one attribute of ParamAttr, gru_unit will create ParamAttr as bias_attr. If the Initializer of the bias_attr is not set, the bias - is initialized zero. Default: None. + is initialized zero. The default value is None. activation (str): The activation type for cell (actNode). - Default: 'tanh' + The default value is 'tanh'. gate_activation (str): The activation type for gates (actGate). - Default: 'sigmoid' - dtype(str): The dtype of the layers. Default: 'float32' + The default value is 'sigmoid'. + dtype(str): The dtype of the layers. The data type can be set as + 'float32', 'float64'. The default value is 'float32'. - Attributes: - weight (Parameter): the learnable weights of this layer. - bias (Parameter): the learnable bias of this layer. + Attribute: + **weight** (Parameter): the learnable weights of this layer. + + **bias** (Parameter): the learnable bias of this layer. Returns: - tuple: The hidden value, reset-hidden value and gate values. + tuple: The hidden value, reset-hidden value and gate values. The hidden value + is a 2-D tensor with shape :math:`[T, D]` . The reset-hidden value is a + 2-D tensor with shape :math:`[T, D]` . The gate value is a 2-D tensor with + shape :math:`[T, 3*D]`. Examples: @@ -1715,6 +1752,7 @@ class GRUUnit(layers.Layer): D = 5 T = sum(lod[0]) + input = numpy.random.rand(T, 3 * D).astype('float32') hidden_input = numpy.random.rand(T, D).astype('float32') with fluid.dygraph.guard(): x = numpy.random.random((3, 32, 32)).astype('float32') @@ -2154,22 +2192,24 @@ class BilinearTensorProduct(layers.Layer): - :math:`y`: the second input contains N elements, shape is [batch_size, N]. - :math:`W_{i}`: the i-th learned weight, shape is [M, N] - :math:`out_{i}`: the i-th element of out, shape is [batch_size, size]. - - :math:`y^\mathrm{T}`: the transpose of :math:`y_{2}`. + - :math:`y^\mathrm{T}`: the transpose of :math:`y`. Parameters: name_scope(str): The name of this class. size (int): The dimension of this layer. - act (str): Activation to be applied to the output of this layer. Default: None. - name (str): The name of this layer. Default: None. - param_attr (ParamAttr): The parameter attribute for the learnable w. - parameters/weights of this layer. Default: None. - bias_attr (ParamAttr): The parameter attribute for the bias + name (str): The default value is None. Normally there is no need for user + to set this property. For more information, please refer to :ref:`api_guide_Name`. + act (str, optional): Activation to be applied to the output of this layer. The default value is None. + param_attr (ParamAttr, optional): The parameter attribute for the learnable w, parameters/weights of + this layer. The default value is None. + bias_attr (ParamAttr, optional): The parameter attribute for the bias of this layer. If it is set to False, no bias will be added to the output units. - If it is set to None, the bias is initialized zero. Default: None. + If it is set to None, the bias is initialized zero. The default value is None. - Attributes: - weight (Parameter): the learnable weights of this layer. - bias (Parameter|None): the learnable bias of this layer. + Attribute: + **weight** (Parameter): the learnable weights of this layer. + + **bias** (Parameter): the learnable bias of this layer. Returns: Variable: A 2-D Tensor of shape [batch_size, size]. diff --git a/python/paddle/fluid/layers/nn.py b/python/paddle/fluid/layers/nn.py index 5fa7b0933f..03d2c46657 100755 --- a/python/paddle/fluid/layers/nn.py +++ b/python/paddle/fluid/layers/nn.py @@ -8986,8 +8986,8 @@ def label_smooth(label, dtype="float32", name=None): """ - Label smoothing is a mechanism to regularize the classifier layer and is - called label-smoothing regularization (LSR). + Label smoothing is a mechanism to regularize the classifier layer and is called + label-smoothing regularization (LSR). Label smoothing is proposed to encourage the model to be less confident, since optimizing the log-likelihood of the correct label directly may @@ -9006,19 +9006,23 @@ def label_smooth(label, See more details about label smoothing in https://arxiv.org/abs/1512.00567. - Args: + Parameters: label(Variable): The input variable containing the label data. The - label data should use one-hot representation. - prior_dist(Variable): The prior distribution to be used to smooth - labels. If not provided, an uniform distribution - is used. The shape of :attr:`prior_dist` should - be :math:`(1, class\_num)`. - epsilon(float): The weight used to mix up the original ground-truth - distribution and the fixed distribution. - dtype(np.dtype|core.VarDesc.VarType|str): The type of data : float32, - float_64, int etc. - name(str|None): A name for this layer(optional). If set None, the layer - will be named automatically. + label data should use one-hot representation. It's + a multidimensional tensor with a shape of + :math:`[N_1, ..., Depth]`, where Depth is class number. + prior_dist(Variable, optional): The prior distribution to be used to smooth + labels. If not provided, an uniform distribution + is used. It's a multidimensional tensor with a shape of + :math:`[1, class\_num]` . The default value is None. + epsilon(float, optional): The weight used to mix up the original ground-truth + distribution and the fixed distribution. The default value is + 0.1. + dtype(np.dtype|core.VarDesc.VarType|str, optional): The data type can be set + as 'float32', 'float64'. The default value is 'float32'. + name(str, optional): The default value is None. Normally there is no need for user + to set this property. For more information, please refer to + :ref:`api_guide_Name`. Returns: Variable: The tensor variable containing the smoothed labels. -- GitLab