update documentation

docs/api/io.rst

@@ -351,7 +351,7 @@ libmxnet data providers
 
 
    :param prefetch_buffer: Backend Param: Number of prefetched parameters
-   :type prefetch_buffer: , optional, default=4
+   :type prefetch_buffer: long (non-negative), optional, default=4
 
 
    :param rand_crop: Augmentation Param: Whether to random crop on the image
@@ -514,7 +514,7 @@ libmxnet data providers
 
 
    :param prefetch_buffer: Backend Param: Number of prefetched parameters
-   :type prefetch_buffer: , optional, default=4
+   :type prefetch_buffer: long (non-negative), optional, default=4
 
    :return: the constructed :class:`MXDataProvider`.
 

docs/api/ndarray.rst

@@ -687,13 +687,54 @@ Public APIs
 
 
 
+
+.. function:: sum_mid_internal(...)
+
+   Take sum on medium dimension of the 3D src.
+
+   :param src: Source input to the function
+   :type src: NDArray
+
+
+
+
+
+.. function:: transpose(...)
+
+   Transpose the input matrix and return a new one
+
+   :param src: Source input to the function
+   :type src: NDArray
+
+
+
+
 Internal APIs
 ^^^^^^^^^^^^^
 
 .. note::
 
    Document and signatures for internal API functions might be incomplete.
 
+.. function:: _broadcast(...)
+
+   Broadcast array in the given axis to the given size
+
+   :param src: source ndarray
+   :type src: NDArray
+
+
+   :param axis: axis to broadcast
+   :type axis: int
+
+
+   :param size: size of broadcast
+   :type size: int
+
+
+
+
+
 .. function:: _copyto(...)
 
 

docs/api/symbolic-node.rst

@@ -346,11 +346,15 @@ Public APIs
 
 .. function:: Crop(...)
 
-   Crop the 2nd and 3rd dim of input data, with the corresponding size of w_h or with width and height of the second input symbol
+   Crop the 2nd and 3rd dim of input data, with the corresponding size of h_w or with width and height of the second input symbol, i.e., with one input, we need h_w to specify the crop height and width, otherwise the second input symbol's size will be used
 
    This function support variable length positional :class:`SymbolicNode` inputs.
 
-   :param num_args: Number of inputs for crop, if equals one, then we will use the h_wfor crop heihgt and width, else if equals two, then we will use the heightand width of the second input symbol, we name crop_like here
+   :param data: Tensor or List of Tensors, the second input will be used as crop_like shape reference
+   :type data: SymbolicNode or SymbolicNode[]
+
+
+   :param num_args: Number of inputs for crop, if equals one, then we will use the h_wfor crop height and width, else if equals two, then we will use the heightand width of the second input symbol, we name crop_like here
    :type num_args: int, required
 
 
@@ -374,6 +378,34 @@ Public APIs
 
 
 
+.. function:: CuDNNBatchNorm(...)
+
+   Apply batch normalization to input.
+
+   :param data: Input data to batch normalization
+   :type data: SymbolicNode
+
+
+   :param eps: Epsilon to prevent div 0
+   :type eps: float, optional, default=0.001
+
+
+   :param momentum: Momentum for moving average
+   :type momentum: float, optional, default=0.9
+
+
+   :param fix_gamma: Fix gamma while training
+   :type fix_gamma: boolean, optional, default=False
+
+   :param Symbol name: The name of the :class:`SymbolicNode`. (e.g. `:my_symbol`), optional.
+   :param Dict{Symbol, AbstractString} attrs: The attributes associated with this :class:`SymbolicNode`.
+
+   :return: SymbolicNode.
+
+
+
+
+
 .. function:: Deconvolution(...)
 
    Apply deconvolution to input then add a bias.
@@ -759,11 +791,15 @@ Public APIs
 
 .. function:: SliceChannel(...)
 
-   Slice channel into many outputs with equally divided channel
+   Slice input equally along specified axis
 
    :param num_outputs: Number of outputs to be sliced.
    :type num_outputs: int, required
 
+
+   :param axis: Dimension along which to slice.
+   :type axis: int, optional, default='1'
+
    :param Symbol name: The name of the :class:`SymbolicNode`. (e.g. `:my_symbol`), optional.
    :param Dict{Symbol, AbstractString} attrs: The attributes associated with this :class:`SymbolicNode`.