Formatting fixes

python/mxnet/image.py

@@ -170,7 +170,7 @@ def fixed_crop(src, x0, y0, w, h, size=None, interp=2):
 
 def random_crop(src, size, interp=2):
     """Randomly crop `src` with `size` (width, height).
-       Upsample result if `src` is smaller than `size`.
+    Upsample result if `src` is smaller than `size`.
 
     Parameters
     ----------
@@ -239,7 +239,7 @@ def center_crop(src, size, interp=2):
 
         4: Lanczos interpolation over 8x8 pixel neighborhood.
 
-         When shrinking an image, it will generally look best with AREA-based
+        When shrinking an image, it will generally look best with AREA-based
         interpolation, whereas, when enlarging an image, it will generally look best
         with Bicubic (slow) or Bilinear (faster but still looks OK).
 
@@ -708,7 +708,6 @@ def read_image(self, fname):
         Example usage:
         ----------
         >>> dataIter.read_image('Face.jpg') # returns decoded raw bytes.
-        '\xff\xd8\xff\xe0\x00...'
         """
         with open(os.path.join(self.path_root, fname), 'rb') as fin:
             img = fin.read()

python/mxnet/io.py

@@ -25,7 +25,7 @@ class DataDesc(namedtuple('DataDesc', ['name', 'shape'])):
     that the first axis is number of examples in the batch(N),
     C is number of channels, H is the height and W is the width of the image.
 
-    for sequential data, by default `layout` is set to ``NTC`` where
+    For sequential data, by default `layout` is set to ``NTC``, where
     N is number of examples in the batch, T the temporal axis representing time
     and C is the number of channels.
 

python/mxnet/symbol.py

@@ -45,7 +45,7 @@ class Symbol(SymbolBase):
     __slots__ = []
 
     def __repr__(self):
-        """Get a string representation of the symbol."""
+        """Gets a string representation of the symbol."""
         name = self.name
         if name is None:
             name = ', '.join([i.name for i in self])
@@ -453,7 +453,7 @@ def __getitem__(self, index):
 
     @property
     def name(self):
-        """Get name string from the symbol, this function only works for non-grouped symbol.
+        """Gets name string from the symbol, this function only works for non-grouped symbol.
 
         Returns
         -------
@@ -1412,7 +1412,7 @@ def bind(self, ctx, args, args_grad=None, grad_req='write',
         return executor
 
     def grad(self, wrt):
-        """Get the autodiff of current symbol.
+        """Gets the autodiff of current symbol.
 
         This function can only be used if current symbol is a loss function.
 
@@ -1793,9 +1793,9 @@ def minimum(left, right):
 # pylint: disable=no-member
 # pylint: disable=redefined-builtin
 def hypot(left, right):
-    """Given the "legs" of a right triangle, return its hypotenuse.
+    """Given the "legs" of a right triangle, returns its hypotenuse.
 
-    Equivalent to "sqrt(left**2 + right**2)", element-wise.
+    Equivalent to :math:`\sqrt(left^2 + right^2)`, element-wise.
     Both inputs can be Symbol or scalar number. Broadcasting is not supported.
 
     Parameters
@@ -1836,7 +1836,7 @@ def hypot(left, right):
 
 
 def zeros(shape, dtype=None, **kwargs):
-    """Return a new symbol of given shape and type, filled with zeros.
+    """Returns a new symbol of given shape and type, filled with zeros.
 
     Parameters
     ----------
@@ -1856,7 +1856,7 @@ def zeros(shape, dtype=None, **kwargs):
 
 
 def ones(shape, dtype=None, **kwargs):
-    """Return a new symbol of given shape and type, filled with ones.
+    """Returns a new symbol of given shape and type, filled with ones.
 
     Parameters
     ----------
@@ -1876,7 +1876,7 @@ def ones(shape, dtype=None, **kwargs):
 
 
 def arange(start, stop=None, step=1.0, repeat=1, name=None, dtype=None):
-    """Return evenly spaced values within a given interval.
+    """Returns evenly spaced values within a given interval.
 
     Parameters
     ----------

src/operator/concat.cc

@@ -51,7 +51,7 @@ MXNET_REGISTER_OP_PROPERTY(Concat, ConcatProp)
 .. note:: `Concat` is deprecated. Use `concat` instead.
 
 The dimensions of the input arrays should be the same except the axis along
- which they will concatenated.
+which they will concatenated.
 The dimension of the output array along the concatenated axis will be equal
 to the sum of the corresponding dimensions of the input arrays.
 

src/operator/loss_binary_op.cc

@@ -25,7 +25,7 @@ NNVM_REGISTER_OP(softmax_cross_entropy)
 
   .. math:: \text{CE(label, output)} = - \sum_i \text{label}_i \log(\text{output}_i)
 
-  Example::
+Example::
 
   x = [[1, 2, 3],
        [11, 7, 5]]

src/operator/slice_channel.cc

@@ -72,6 +72,8 @@ Example::
 along the `axis` which it is split.
 Also `squeeze_axis` can be set to true only if ``input.shape[axis] == num_outputs``.
 
+Example::
+
    z = split(x, axis=0, num_outputs=3, squeeze_axis=1) // a list of 3 arrays with shape (2, 1)
    z = [[ 1.]
         [ 2.]]

src/operator/softmax_output.cc

@@ -72,7 +72,7 @@ MXNET_REGISTER_OP_PROPERTY(SoftmaxOutput, SoftmaxOutputProp)
 
   - If the parameter `use_ignore` is ``true``, `ignore_label` can specify input instances
     with a particular label to be ignored during backward propagation. **This has no effect when
-    softmax** `output` **has same shape as** `label`.
+    softmax `output` has same shape as `label`**.
 
     Example::
 

src/operator/tensor/indexing_op.cc

@@ -189,7 +189,7 @@ The locations represented by `indices` take value `on_value`, while all
 other locations take value `off_value`.
 
 `one_hot` operation with `indices` of shape ``(i0, i1)`` and `depth`  of ``d`` would result
- in an output array of shape ``(i0, i1, d)`` with::
+in an output array of shape ``(i0, i1, d)`` with::
 
   output[i,j,:] = off_value
   output[i,j,indices[i,j]] = on_value

src/operator/tensor/matrix_op.cc

@@ -230,7 +230,7 @@ and ``end=(e_1, e_2, ... e_n)`` indices will result in an array with the shape
 ``(e_1-b_0, ..., e_n-b_n-1)``.
 
 The resulting array's *k*-th dimension contains elements
- from the *k*-th dimension of the input array with the open range ``[b_k, e_k)``.
+from the *k*-th dimension of the input array with the open range ``[b_k, e_k)``.
 
 Example::
 

src/operator/tensor/multisample_op.cc

@@ -155,13 +155,15 @@ DMLC_REGISTER_PARAMETER(MultiSampleParam);
 inline std::string uniform_desc() {
   return std::string(R"code(Concurrent sampling from multiple 
 uniform distributions on the intervals given by *[low,high)*.
+
 The parameters of the distributions are provided as input arrays.
 Let *[s]* be the shape of the input arrays, *n* be the dimension of *[s]*, *[t]*
 be the shape specified as the parameter of the operator, and *m* be the dimension
-of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*. For any
-valid *n*-dimensional index *i* with respect to the input arrays, *output[i]* will be
-an *m*-dimensional array that holds randomly drawn samples from the distribution which
-is parameterized by the input values at index *i*. If the shape parameter of the
+of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*.
+
+For any valid *n*-dimensional index *i* with respect to the input arrays, *output[i]*
+will be an *m*-dimensional array that holds randomly drawn samples from the distribution
+which is parameterized by the input values at index *i*. If the shape parameter of the
 operator is not set, then one sample will be drawn per distribution and the output array
 has the same shape as the input arrays.
 
@@ -182,13 +184,15 @@ Examples::
 inline std::string normal_desc() {
   return std::string(R"code(Concurrent sampling from multiple 
 normal distributions with parameters *mu* (mean) and *sigma* (standard deviation).
+
 The parameters of the distributions are provided as input arrays.
 Let *[s]* be the shape of the input arrays, *n* be the dimension of *[s]*, *[t]*
 be the shape specified as the parameter of the operator, and *m* be the dimension
-of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*. For any
-valid *n*-dimensional index *i* with respect to the input arrays, *output[i]* will be
-an *m*-dimensional array that holds randomly drawn samples from the distribution which
-is parameterized by the input values at index *i*. If the shape parameter of the
+of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*.
+
+For any valid *n*-dimensional index *i* with respect to the input arrays, *output[i]*
+will be an *m*-dimensional array that holds randomly drawn samples from the distribution
+which is parameterized by the input values at index *i*. If the shape parameter of the
 operator is not set, then one sample will be drawn per distribution and the output array
 has the same shape as the input arrays.
 
@@ -209,13 +213,15 @@ Examples::
 inline std::string gamma_desc() {
   return std::string(R"code(Concurrent sampling from multiple 
 gamma distributions with parameters *alpha* (shape) and *beta* (scale).
+
 The parameters of the distributions are provided as input arrays.
 Let *[s]* be the shape of the input arrays, *n* be the dimension of *[s]*, *[t]*
 be the shape specified as the parameter of the operator, and *m* be the dimension
-of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*. For any
-valid *n*-dimensional index *i* with respect to the input arrays, *output[i]* will be
-an *m*-dimensional array that holds randomly drawn samples from the distribution which
-is parameterized by the input values at index *i*. If the shape parameter of the
+of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*.
+
+For any valid *n*-dimensional index *i* with respect to the input arrays, *output[i]*
+will be an *m*-dimensional array that holds randomly drawn samples from the distribution
+which is parameterized by the input values at index *i*. If the shape parameter of the
 operator is not set, then one sample will be drawn per distribution and the output array
 has the same shape as the input arrays.
 
@@ -236,13 +242,15 @@ Examples::
 inline std::string exponential_desc() {
   return std::string(R"code(Concurrent sampling from multiple 
 exponential distributions with parameters lambda (rate).
+
 The parameters of the distributions are provided as an input array.
 Let *[s]* be the shape of the input array, *n* be the dimension of *[s]*, *[t]*
 be the shape specified as the parameter of the operator, and *m* be the dimension
-of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*. For any
-valid *n*-dimensional index *i* with respect to the input array, *output[i]* will be
-an *m*-dimensional array that holds randomly drawn samples from the distribution which
-is parameterized by the input value at index *i*. If the shape parameter of the
+of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*.
+
+For any valid *n*-dimensional index *i* with respect to the input array, *output[i]*
+will be an *m*-dimensional array that holds randomly drawn samples from the distribution
+which is parameterized by the input value at index *i*. If the shape parameter of the
 operator is not set, then one sample will be drawn per distribution and the output array
 has the same shape as the input array.
 
@@ -262,15 +270,18 @@ Examples::
 inline std::string poisson_desc() {
   return std::string(R"code(Concurrent sampling from multiple 
 Poisson distributions with parameters lambda (rate).
+
 The parameters of the distributions are provided as an input array.
 Let *[s]* be the shape of the input array, *n* be the dimension of *[s]*, *[t]*
 be the shape specified as the parameter of the operator, and *m* be the dimension
-of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*. For any
-valid *n*-dimensional index *i* with respect to the input array, *output[i]* will be
-an *m*-dimensional array that holds randomly drawn samples from the distribution which
-is parameterized by the input value at index *i*. If the shape parameter of the
+of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*.
+
+For any valid *n*-dimensional index *i* with respect to the input array, *output[i]*
+will be an *m*-dimensional array that holds randomly drawn samples from the distribution
+which is parameterized by the input value at index *i*. If the shape parameter of the
 operator is not set, then one sample will be drawn per distribution and the output array
 has the same shape as the input array.
+
 Samples will always be returned as a floating point data type.
 
 Examples::
@@ -289,15 +300,18 @@ Examples::
 inline std::string negative_binomial_desc() {
   return std::string(R"code(Concurrent sampling from multiple 
 negative binomial distributions with parameters *k* (failure limit) and *p* (failure probability).
+
 The parameters of the distributions are provided as input arrays.
 Let *[s]* be the shape of the input arrays, *n* be the dimension of *[s]*, *[t]*
 be the shape specified as the parameter of the operator, and *m* be the dimension
-of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*. For any
-valid *n*-dimensional index *i* with respect to the input arrays, *output[i]* will be
-an *m*-dimensional array that holds randomly drawn samples from the distribution which
-is parameterized by the input values at index *i*. If the shape parameter of the
+of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*.
+
+For any valid *n*-dimensional index *i* with respect to the input arrays, *output[i]*
+will be an *m*-dimensional array that holds randomly drawn samples from the distribution
+which is parameterized by the input values at index *i*. If the shape parameter of the
 operator is not set, then one sample will be drawn per distribution and the output array
 has the same shape as the input arrays.
+
 Samples will always be returned as a floating point data type.
 
 Examples::
@@ -317,15 +331,18 @@ Examples::
 inline std::string generalized_negative_binomial_desc() {
   return std::string(R"code(Concurrent sampling from multiple 
 generalized negative binomial distributions with parameters *mu* (mean) and *alpha* (dispersion).
+
 The parameters of the distributions are provided as input arrays.
 Let *[s]* be the shape of the input arrays, *n* be the dimension of *[s]*, *[t]*
 be the shape specified as the parameter of the operator, and *m* be the dimension
-of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*. For any
-valid *n*-dimensional index *i* with respect to the input arrays, *output[i]* will be
-an *m*-dimensional array that holds randomly drawn samples from the distribution which
-is parameterized by the input values at index *i*. If the shape parameter of the
+of *[t]*. Then the output will be a *(n+m)*-dimensional array with shape *[s]x[t]*.
+
+For any valid *n*-dimensional index *i* with respect to the input arrays, *output[i]*
+will be an *m*-dimensional array that holds randomly drawn samples from the distribution
+which is parameterized by the input values at index *i*. If the shape parameter of the
 operator is not set, then one sample will be drawn per distribution and the output array
 has the same shape as the input arrays.
+
 Samples will always be returned as a floating point data type.
 
 Examples::

src/operator/upsampling.cc

@@ -54,7 +54,7 @@ Operator* UpSamplingProp::CreateOperatorEx(Context ctx, std::vector<TShape> *in_
 DMLC_REGISTER_PARAMETER(UpSamplingParam);
 
 MXNET_REGISTER_OP_PROPERTY(UpSampling, UpSamplingProp)
-.describe("Performs nearest neighbor/bilinear up sampling to inputs")
+.describe("Performs nearest neighbor/bilinear up sampling to inputs.")
 .add_argument("data", "NDArray-or-Symbol[]", "Array of tensors to upsample")
 .add_arguments(UpSamplingParam::__FIELDS__())
 .set_key_var_num_args("num_args");