本文共 29718 字，大约阅读时间需要 99 分钟。

Table of Contents

---

 

Members
	for 4-D tensors of type T.
	for N-D tensors of type T.
	Bitcasts a tensor from one type to another without copying data.
	Return the shape of s0 op s1 with broadcast.
	Broadcast an array for a compatible shape.
	Checks a tensor for NaN and Inf values.
	Concatenates tensors along one dimension.
	Shuffle dimensions of x according to a permutation and conjugate the result.
	op for gradient debugging.
	op for gradient debugging.
	Makes a copy of x.
	for tensors of type T.
	the 'input' tensor into a float .
	Returns a diagonal tensor with a given diagonal values.
	Returns the diagonal part of the tensor.
	Computes the (possibly normalized) Levenshtein Edit Distance.
	Creates a tensor with the given shape.
	Ensures that the tensor's shape matches the expected shape.
	Inserts a dimension of 1 into a tensor's shape.
	Extract patches from images and put them in the "depth" output dimension.
	Extract patches from input and put them in the "depth" output dimension.
	Fake-quantize the 'inputs' tensor, type float to 'outputs' tensor of same type.
	Compute gradients for a  operation.
	Fake-quantize the 'inputs' tensor of type float via global float scalars min
	Compute gradients for a  operation.
	Fake-quantize the 'inputs' tensor of type float and one of the shapes: [d],.
	Compute gradients for a operation.
	Creates a tensor filled with a scalar value.
	slices from params according to indices.
	slices from params into a  with shape specified by indices.
	slices from params axis axis according to indices.
	Gives a guarantee to the TF runtime that the input tensor is a constant.
	Return a tensor with the same shape and contents as the input tensor or value.
	Returns a list of tensors with the same shapes and contents as the input.
	Returns immutable tensor from memory region.
	Adds v into specified rows of x.
	Subtracts v into specified rows of x.
	Updates specified rows with values in v.
	Computes the inverse permutation of a tensor.
	Copy a tensor setting everything outside a central band in each innermost matrix.
	Returns a batched diagonal tensor with a given batched diagonal values.
	Returns the batched diagonal part of a batched tensor.
	Returns a batched matrix tensor with new batched diagonal values.
	Pads a tensor with mirrored values.
	Returns a one-hot tensor.
	Returns a tensor of ones with the same shape and type as x.
	Packs a list of N rank-R tensors into one rank-(R+1) tensor.
	Returns the rank of a tensor.
	Reshapes a tensor.
	value to the sliced l-value reference of ref.
	Reverses specific dimensions of a tensor.
	Reverses variable length slices.
	Scatter updates into a new tensor according to indices.
	Applies sparse addition to input using individual values or slices.
	Computes the difference between two lists of numbers or strings.
	Returns the shape of a tensor.
	Returns shape of tensors.
	Returns the size of a tensor.
	Return a slice from 'input'.
	Returns a copy of the input tensor.
	for 4-D tensors of type T.
	for N-D tensors of type T.
	for tensors of type T.
	Splits a tensor into num_split tensors along one dimension.
	Splits a tensor into num_split tensors along one dimension.
	Removes dimensions of size 1 from the shape of a tensor.
	Stops gradient computation.
	Return a strided slice from input.
	value to the sliced l-value reference of ref.
	Returns the gradient of .
	Constructs a tensor by tiling a given tensor.
	Shuffle dimensions of x according to a permutation.
	Finds unique elements in a 1-D tensor.
	Finds unique elements along an axis of a tensor.
	Finds unique elements in a 1-D tensor.
	Finds unique elements along an axis of a tensor.
	Converts a flat index or array of flat indices into a tuple of.
	Unpacks a given dimension of a rank-R tensor into num rank-(R-1)tensors.
	Returns locations of nonzero / true values in a tensor.
	Returns a tensor of zeros with the same shape and type as x.

 

Members
	Generates labels for candidate sampling with a learned unigram distribution.
	Computes the ids of the positions in sampled_candidates that match true_labels.
	Generates labels for candidate sampling with a learned unigram distribution.
	Generates labels for candidate sampling with a learned unigram distribution.
	Generates labels for candidate sampling with a log-uniform distribution.
	Generates labels for candidate sampling with a uniform distribution.

 

Members
	Raise a exception to abort the process when called.
	Does nothing.
	Forwards the input to the output.
	Forwards the value of an available tensor from inputs to output.
	Makes its input available to the next iteration.
	Makes its input available to the next iteration.
	Forwards the indexth element of inputs to output.
	Forwards the ref tensor data to the output port determined by pred.
	Forwards data to the output port determined by pred.

 

Members
	A  object lets the caller drive the evaluation of the TensorFlow graph constructed with the C++ API.
	Represents a tensor value that can be used as an operand to an .
	A type for representing the input to ops that require a list of tensors.
	Represents a node in the computation graph.
	Represents a tensor value produced by an .
	A  object represents a set of related TensorFlow ops that have the same properties such as a common name prefix.
	Denotes success or failure of a call in Tensorflow.
	Represents an n-dimensional array of values.

 

Members
	Applies a gradient to a given accumulator.
	Returns the number of gradients aggregated in the given accumulators.
	Updates the accumulator with a new value for global_step.
	Extracts the average gradient in the given .
	Defines a barrier that persists across different graph executions.
	Closes the given barrier.
	Computes the number of incomplete elements in the given barrier.
	For each key, assigns the respective value to the specified component.
	Computes the number of complete elements in the given barrier.
	Takes the given number of completed elements from a barrier.
	A conditional accumulator for aggregating gradients.
	Delete the tensor specified by its handle in the session.
	Partitions data into num_partitions tensors using indices from partitions.
	Interleave the values from the data tensors into a single tensor.
	A queue that produces elements in first-in first-out order.
	Store the input tensor in the state of the current session.
	Store the input tensor in the state of the current session.
	Get the value of the tensor specified by its handle.
	Op removes all elements in the underlying container.
	Op returns the number of incomplete elements in the underlying container.
	Op peeks at the values at the specified key.
	Op returns the number of elements in the underlying container.
	(key, values) in the underlying container which behaves like a hashtable.
	Op removes and returns the values associated with the key.
	Op removes and returns a random (key, value)
	Op removes all elements in the underlying container.
	Op returns the number of incomplete elements in the underlying container.
	Op peeks at the values at the specified key.
	Op returns the number of elements in the underlying container.
	(key, values) in the underlying container which behaves like a ordered.
	Op removes and returns the values associated with the key.
	Op removes and returns the (key, value) element with the smallest.
	A queue that produces elements in first-in first-out order.
	Interleave the values from the data tensors into a single tensor.
	A queue that produces elements sorted by the first component value.
	Closes the given queue.
	Dequeues a tuple of one or more tensors from the given queue.
	Dequeues n tuples of one or more tensors from the given queue.
	Dequeues n tuples of one or more tensors from the given queue.
	Enqueues a tuple of one or more tensors in the given queue.
	Enqueues zero or more tuples of one or more tensors in the given queue.
	Returns true if queue is closed.
	Returns true if queue is closed.
	Computes the number of elements in the given queue.
	A queue that randomizes the order of elements.
	Emits randomized records.
	Applies a sparse gradient to a given accumulator.
	Extracts the average sparse gradient in a .
	A conditional accumulator for aggregating sparse gradients.
	values similar to a lightweight Enqueue.
	Op removes all elements in the underlying container.
	Op peeks at the values at the specified index.
	Op returns the number of elements in the underlying container.
	An array of Tensors of given size.
	Delete the  from its resource container.
	the elements from the  into value value.
	specific elements from the  into output value.
	Creates a  for storing the gradients of values in the given handle.
	Creates a  for storing multiple gradients of values in the given handle.
	Read an element from the  into output value.
	Scatter the data from the input value into specific  elements.
	Get the current size of the .
	the data from the input value into  elements.
	Push an element onto the tensor_array.
	Op is similar to a lightweight Dequeue.

 

Members
	Adjust the contrast of one or more images.
	Adjust the hue of one or more images.
	Adjust the saturation of one or more images.
	Extracts crops from the input image tensor and resizes them.
	Computes the gradient of the crop_and_resize op wrt the input boxes tensor.
	Computes the gradient of the crop_and_resize op wrt the input image tensor.
	Decode and Crop a JPEG-encoded image to a uint8 tensor.
	Decode the first frame of a BMP-encoded image to a uint8 tensor.
	Decode the first frame of a GIF-encoded image to a uint8 tensor.
	Decode a JPEG-encoded image to a uint8 tensor.
	Decode a PNG-encoded image to a uint8 or uint16 tensor.
	Draw bounding boxes on a batch of images.
	JPEG-encode an image.
	PNG-encode an image.
	Extracts a glimpse from the input tensor.
	Extract the shape information of a JPEG-encoded image.
	Convert one or more images from HSV to RGB.
	Greedily selects a subset of bounding boxes in descending order of score,.
	Greedily selects a subset of bounding boxes in descending order of score,.
	Greedily selects a subset of bounding boxes in descending order of score,.
	Greedily selects a subset of bounding boxes in descending order of score,.
	Greedily selects a subset of bounding boxes in descending order of score,.
	Resize quantized images to size using quantized bilinear interpolation.
	Converts one or more images from RGB to HSV.
	Resize images to size using area interpolation.
	Resize images to size using bicubic interpolation.
	Resize images to size using bilinear interpolation.
	Resize images to size using nearest neighbor interpolation.
	Generate a single randomly distorted bounding box for an image.
	Generate a single randomly distorted bounding box for an image.

 

Members
	A Reader that outputs fixed-length records from a file.
	A Reader that outputs the queued work as both the key and value.
	A Reader that outputs the records from a LMDB file.
	Returns the set of files matching one or more glob patterns.
	V2 format specific: merges the metadata files of sharded checkpoints.
	Reads and outputs the entire contents of the input filename.
	Returns the number of records this Reader has produced.
	Returns the number of work units this Reader has finished processing.
	Returns the next record (key, value pair) produced by a Reader.
	Returns up to num_records (key, value) pairs produced by a Reader.
	a Reader to its initial clean state.
	a reader to a previously saved state.
	Produce a string tensor that encodes the state of a Reader.
	Restores a tensor from checkpoint files.
	Restores a tensor from checkpoint files.
	Restores tensors from a V2 checkpoint.
	Saves the input tensors to disk.
	Saves input tensors slices to disk.
	Saves tensors in V2 checkpoint format.
	Generate a sharded filename.
	Generate a glob pattern matching all sharded file names.
	A Reader that outputs the records from a TensorFlow Records file.
	A Reader that outputs the lines of a file delimited by ' '.
	A Reader that outputs the entire contents of a file as a value.
	Writes contents to the file at input filename.

 

Members
	Asserts that the given condition is true.
	Outputs a Summary protocol buffer with a histogram.
	Merges summaries.
	Prints a list of tensors.
	Prints a string scalar.
	Outputs a Summary protocol buffer with scalar values.
	Outputs a Summary protocol buffer with a tensor.
	Outputs a Summary protocol buffer with a tensor and per-plugin data.
	Provides the time since epoch in seconds.

 

Members
	Computes the absolute value of a tensor.
	Returns the element-wise sum of a list of tensors.
	Computes acos of x element-wise.
	Computes inverse hyperbolic cosine of x element-wise.
	Returns x + y element-wise.
	all input tensors element wise.
	Returns x + y element-wise.
	Computes the "logical and" of elements across dimensions of a tensor.
	Returns the argument of a complex number.
	Computes the "logical or" of elements across dimensions of a tensor.
	Returns the truth value of abs(x-y) < tolerance element-wise.
	Returns the index with the largest value across dimensions of a tensor.
	Returns the index with the smallest value across dimensions of a tensor.
	Computes asin of x element-wise.
	Computes inverse hyperbolic sine of x element-wise.
	Computes atan of x element-wise.
	Computes arctangent of y/x element-wise, respecting signs of the arguments.
	Computes inverse hyperbolic tangent of x element-wise.
	Multiplies slices of two tensors in batches.
	Computes the Bessel i0e function of x element-wise.
	Computes the Bessel i1e function of x element-wise.
	Compute the regularized incomplete beta integral \(I_x(a, b)\).
	Counts the number of occurrences of each value in an integer array.
	Bucketizes 'input' based on 'boundaries'.
	x of type SrcT to y of DstT.
	Returns element-wise smallest integer not less than x.
	Clips tensor values to a specified min and max.
	Compare values of input to threshold and pack resulting bits into a uint8.
	Converts two real numbers to a complex number.
	Computes the complex absolute value of a tensor.
	Returns the complex conjugate of a complex number.
	Computes cos of x element-wise.
	Computes hyperbolic cosine of x element-wise.
	Compute the pairwise cross product.
	Compute the cumulative product of the tensor x along axis.
	Compute the cumulative sum of the tensor x along axis.
	Computes Psi, the derivative of  (the log of the absolute value of.
	Returns x / y element-wise.
	Returns 0 if the denominator is zero.
	Returns the truth value of (x == y) element-wise.
	Computes the Gauss error function of x element-wise.
	Computes the complementary error function of x element-wise.
	Computes exponential of x element-wise.
	Computes exponential of x - 1 element-wise.
	Returns element-wise largest integer not greater than x.
	Returns x // y element-wise.
	Returns element-wise remainder of division.
	Returns the truth value of (x > y) element-wise.
	Returns the truth value of (x >= y) element-wise.
	Return histogram of values.
	Compute the lower regularized incomplete Gamma function P(a, x).
	Compute the upper regularized incomplete Gamma function Q(a, x).
	Returns the imaginary part of a complex number.
	Computes the reciprocal of x element-wise.
	Returns which elements of x are finite.
	Returns which elements of x are Inf.
	Returns which elements of x are NaN.
	Returns the truth value of (x < y) element-wise.
	Returns the truth value of (x <= y) element-wise.
	Computes the log of the absolute value of Gamma(x) element-wise.
	Generates values in an interval.
	Computes natural logarithm of x element-wise.
	Computes natural logarithm of (1 + x) element-wise.
	Returns the truth value of x AND y element-wise.
	Returns the truth value of NOT x element-wise.
	Returns the truth value of x OR y element-wise.
	the matrix "a" by the matrix "b".
	Computes the maximum of elements across dimensions of a tensor.
	Returns the max of x and y (i.e.
	Computes the mean of elements across dimensions of a tensor.
	Computes the minimum of elements across dimensions of a tensor.
	Returns the min of x and y (i.e.
	Returns element-wise remainder of division.
	Returns x * y element-wise.
	Computes numerical negative value element-wise.
	Returns the truth value of (x != y) element-wise.
	Compute the polygamma function \(^{(n)}(x)\).
	Computes the power of one value to another.
	Computes the product of elements across dimensions of a tensor.
	Convert the quantized 'input' tensor into a lower-precision 'output', using the.
	Returns x + y element-wise, working on quantized buffers.
	Perform a quantized matrix multiplication of a by the matrix b.
	Returns x * y element-wise, working on quantized buffers.
	Creates a sequence of numbers.
	Returns the real part of a complex number.
	Returns x / y element-wise for real types.
	Computes the reciprocal of x element-wise.
	Given a quantized tensor described by (input, input_min, input_max), outputs a.
	Convert the quantized 'input' tensor into a lower-precision 'output', using the.
	Returns element-wise integer closest to x.
	Rounds the values of a tensor to the nearest integer, element-wise.
	Computes reciprocal of square root of x element-wise.
	Computes the maximum along segments of a tensor.
	Computes the mean along segments of a tensor.
	Computes the minimum along segments of a tensor.
	Computes the product along segments of a tensor.
	Computes the sum along segments of a tensor.
	Computes sigmoid of x element-wise.
	Returns an element-wise indication of the sign of a number.
	Computes sin of x element-wise.
	Computes hyperbolic sine of x element-wise.
	matrix "a" by matrix "b".
	Computes the mean along sparse segments of a tensor.
	Computes gradients for .
	Computes the mean along sparse segments of a tensor.
	Computes the sum along sparse segments of a tensor divided by the sqrt of N.
	Computes gradients for .
	Computes the sum along sparse segments of a tensor divided by the sqrt of N.
	Computes the sum along sparse segments of a tensor.
	Computes the sum along sparse segments of a tensor.
	Computes square root of x element-wise.
	Computes square of x element-wise.
	Returns (x - y)(x - y) element-wise.
	Returns x - y element-wise.
	Computes the sum of elements across dimensions of a tensor.
	Computes tan of x element-wise.
	Computes hyperbolic tangent of x element-wise.
	Returns x / y element-wise for integer types.
	Returns element-wise remainder of division.
	Computes the maximum along segments of a tensor.
	Computes the minimum along segments of a tensor.
	Computes the product along segments of a tensor.
	Computes the sum along segments of a tensor.
	Selects elements from x or y, depending on condition.
	Returns 0 if x == 0, and x / y otherwise, elementwise.
	Returns 0 if x == 0, and x * log(y) otherwise, elementwise.
	Compute the Hurwitz zeta function \((x, q)\).

 

Members
	Performs average pooling on the input.
	Performs 3D average pooling on the input.
	Computes gradients of average pooling function.
	Adds bias to value.
	The backward operation for "BiasAdd" on the "bias" tensor.
	Computes a 2-D convolution given 4-D input and filter tensors.
	Computes the gradients of convolution with respect to the filter.
	Computes the gradients of convolution with respect to the input.
	Computes a 3-D convolution given 5-D input and filter tensors.
	Computes the gradients of 3-D convolution with respect to the filter.
	Computes the gradients of 3-D convolution with respect to the input.
	Returns the dimension index in the destination data format given the one in.
	Returns the permuted vector/tensor in the destination data format given the.
	Computes a 2-D depthwise convolution given 4-D input and filtertensors.
	Computes the gradients of depthwise convolution with respect to the filter.
	Computes the gradients of depthwise convolution with respect to the input.
	Computes the grayscale dilation of 4-D input and 3-D filtertensors.
	Computes the gradient of morphological 2-D dilation with respect to the filter.
	Computes the gradient of morphological 2-D dilation with respect to the input.
	Computes exponential linear: exp(features) - 1 if < 0, features otherwise.
	Performs fractional average pooling on the input.
	Performs fractional max pooling on the input.
	Batch normalization.
	Gradient for batch normalization.
	Gradient for batch normalization.
	Batch normalization.
	Performs a padding as a preprocess during a convolution.
	Performs a resize and padding as a preprocess during a convolution.
	Says whether the targets are in the top K predictions.
	Says whether the targets are in the top K predictions.
	L2 Loss.
	Local Response Normalization.
	Computes log softmax activations.
	Performs max pooling on the input.
	Performs 3D max pooling on the input.
	Computes gradients of max pooling function.
	Computes second-order gradients of the maxpooling function.
	Computes second-order gradients of the maxpooling function.
	Computes second-order gradients of the maxpooling function.
	Computes second-order gradients of the maxpooling function.
	Computes gradients of the maxpooling function.
	Performs max pooling on the input.
	Performs max pooling on the input and outputs both max values and indices.
	Finds values of the n-th order statistic for the last dimension.
	Produces the average pool of the input tensor for quantized types.
	Quantized Batch normalization.
	Adds  'bias' to  'input' for Quantized types.
	Computes a 2D convolution given quantized 4D input and filter tensors.
	Produces the max pool of the input tensor for quantized types.
	Computes Quantized Rectified Linear: max(features, 0)
	Computes Quantized Rectified Linear 6: min(max(features, 0), 6)
	Computes Quantized Rectified Linear X: min(max(features, 0), max_value)
	Computes rectified linear: max(features, 0).
	Computes rectified linear 6: min(max(features, 0), 6).
	Computes scaled exponential linear: scale * alpha * (exp(features) - 1)
	Computes softmax activations.
	Computes softmax cross entropy cost and gradients to backpropagate.
	Computes softplus: log(exp(features) + 1).
	Computes softsign: features / (abs(features) + 1).
	Computes softmax cross entropy cost and gradients to backpropagate.
	Finds values and indices of the k largest elements for the last dimension.

 

Members
	Does nothing.

 

Members
	Convert CSV records to tensors.
	Decompress strings.
	Convert JSON-encoded Example records to binary protocol buffer strings.
	Reinterpret the bytes of a string as a vector of numbers.
	Transforms a vector of brain.Example protos (as strings) into typed tensors.
	Transforms a vector of brain.SequenceExample protos (as strings) into typed tensors.
	Transforms a tf.Example proto (as a string) into typed tensors.
	Transforms a scalar brain.SequenceExample proto (as strings) into typed tensors.
	Transforms a serialized tensorflow.TensorProto proto into a .
	Transforms a  into a serialized TensorProto proto.
	Converts each string in the input  to the specified numeric type.

 

Members
	Draws samples from a multinomial distribution.
	Outputs random values from a normal distribution.
	Outputs random values from the Gamma distribution(s) described by alpha.
	Outputs random values from a normal distribution.
	Outputs random values from the Poisson distribution(s) described by rate.
	Randomly shuffles a tensor along its first dimension.
	Outputs random values from a uniform distribution.
	Outputs random integers from a uniform distribution.
	Outputs random values from a truncated normal distribution.

 

Members
	an N-minibatch SparseTensor to a SparseTensorsMap, return N handles.
	a SparseTensor to a SparseTensorsMap return its handle.
	Deserialize and concatenate SparseTensors from a serialized minibatch.
	Deserialize SparseTensor objects.
	Serialize an N-minibatch SparseTensor into an [N, 3] object.
	Serialize a SparseTensor into a [3] object.
	Adds two SparseTensor objects to produce another SparseTensor.
	The gradient operator for the  op.
	Concatenates a list of SparseTensor along the specified dimension.
	Generates sparse cross from a list of sparse and dense tensors.
	Adds up a SparseTensor and a dense , using these special rules:
	Component-wise divides a SparseTensor by a dense .
	Component-wise multiplies a SparseTensor by a dense .
	Fills empty rows in the input 2-D SparseTensor with a default value.
	The gradient of .
	Computes the max of elements across dimensions of a SparseTensor.
	Computes the max of elements across dimensions of a SparseTensor.
	Computes the sum of elements across dimensions of a SparseTensor.
	Computes the sum of elements across dimensions of a SparseTensor.
	Reorders a SparseTensor into the canonical, row-major ordering.
	Reshapes a SparseTensor to represent values in a new dense shape.
	a SparseTensor based on the start and size.
	The gradient operator for the  op.
	Applies softmax to a batched N-D SparseTensor.
	Returns the element-wise max of two SparseTensors.
	Returns the element-wise min of two SparseTensors.
	a SparseTensor into num_split tensors along one dimension.
	Adds up a SparseTensor and a dense , producing a dense .
	SparseTensor (of rank 2) "A" by dense matrix "B".
	Converts a sparse representation into a dense tensor.

 

Members
	Update 'ref' by assigning 'value' to it.
	Update 'ref' by adding 'value' to it.
	Update 'ref' by subtracting 'value' from it.
	Increments 'ref' until it reaches 'limit'.
	Destroys the temporary variable and returns its final value.
	Checks whether a tensor has been initialized.
	Increments variable pointed to by 'resource' until it reaches 'limit'.
	Adds sparse updates to individual values or slices within a given.
	Applies sparse updates to individual values or slices within a given.
	Adds sparse updates to a variable reference.
	Divides a variable reference by sparse updates.
	Reduces sparse updates into a variable reference using the max operation.
	Reduces sparse updates into a variable reference using the min operation.
	Multiplies sparse updates into a variable reference.
	Applies sparse addition between updates and individual values or slices.
	Applies sparse subtraction between updates and individual values or slices.
	Applies sparse updates to individual values or slices within a given.
	Subtracts sparse updates to a variable reference.
	Applies sparse updates to a variable reference.
	Returns a tensor that may be mutated, but only persists within a single step.
	Holds state in the form of a tensor that persists across steps.

 

Members
	Converts each entry in the given tensor to strings.
	Decode web-safe base64-encoded strings.
	Encode strings into web-safe base64 format.
	Joins a string  across the given dimensions.
	Check if the input matches the regex pattern.
	Replaces the match of pattern in input with rewrite.
	Formats a string template using a list of tensors.
	Joins the strings in the given list of string tensors into one tensor;.
	String lengths of input.
	elements of input based on delimiter into a SparseTensor.
	elements of source based on sep into a SparseTensor.
	Strip leading and trailing whitespaces from the .
	Converts each string in the input  to its hash mod by a number of buckets.
	Converts each string in the input  to its hash mod by a number of buckets.
	Converts each string in the input  to its hash mod by a number of buckets.
	Return substrings from  of strings.
	Determine the script codes of a given tensor of Unicode integer code points.

 

Members
	Update '*var' according to the adadelta scheme.
	Update '*var' according to the adagrad scheme.
	Update '*var' according to the proximal adagrad scheme.
	Update '*var' according to the Adam algorithm.
	Update '*var' according to the AddSign update.
	Update '*var' according to the centered RMSProp algorithm.
	Update '*var' according to the Ftrl-proximal scheme.
	Update '*var' according to the Ftrl-proximal scheme.
	Update '*var' by subtracting 'alpha' * 'delta' from it.
	Update '*var' according to the momentum scheme.
	Update '*var' according to the AddSign update.
	Update '*var' and '*accum' according to FOBOS with Adagrad learning rate.
	Update '*var' as FOBOS algorithm with fixed learning rate.
	Update '*var' according to the RMSProp algorithm.
	Update '*var' according to the adadelta scheme.
	Update '*var' according to the adagrad scheme.
	Update '*var' according to the proximal adagrad scheme.
	Update '*var' according to the Adam algorithm.
	Update '*var' according to the AddSign update.
	Update '*var' according to the centered RMSProp algorithm.
	Update '*var' according to the Ftrl-proximal scheme.
	Update '*var' according to the Ftrl-proximal scheme.
	Update '*var' by subtracting 'alpha' * 'delta' from it.
	Update '*var' according to the momentum scheme.
	Update '*var' according to the AddSign update.
	Update '*var' and '*accum' according to FOBOS with Adagrad learning rate.
	Update '*var' as FOBOS algorithm with fixed learning rate.
	Update '*var' according to the RMSProp algorithm.
	var: Should be from a Variable().
	Update relevant entries in '*var' and '*accum' according to the adagrad scheme.
	Update entries in '*var' and '*accum' according to the proximal adagrad scheme.
	Update '*var' according to the centered RMSProp algorithm.
	Update relevant entries in '*var' according to the Ftrl-proximal scheme.
	Update relevant entries in '*var' according to the Ftrl-proximal scheme.
	Update relevant entries in '*var' and '*accum' according to the momentum scheme.
	Sparse update entries in '*var' and '*accum' according to FOBOS algorithm.
	Sparse update '*var' as FOBOS algorithm with fixed learning rate.
	Update '*var' according to the RMSProp algorithm.
	var: Should be from a Variable().
	Update relevant entries in '*var' and '*accum' according to the adagrad scheme.
	Update entries in '*var' and '*accum' according to the proximal adagrad scheme.
	Update '*var' according to the centered RMSProp algorithm.
	Update relevant entries in '*var' according to the Ftrl-proximal scheme.
	Update relevant entries in '*var' according to the Ftrl-proximal scheme.
	Update relevant entries in '*var' and '*accum' according to the momentum scheme.
	Sparse update entries in '*var' and '*accum' according to FOBOS algorithm.
	Sparse update '*var' as FOBOS algorithm with fixed learning rate.
	Update '*var' according to the RMSProp algorithm.

 

Members
	a fact about factorials.

---

转载地址：http://qptaf.baihongyu.com/