From a3120501a3c43da8ec2a0a17008c3b5ee7fcf452 Mon Sep 17 00:00:00 2001
From: lihongkang <[lihongkang1@huawei.com]>
Date: Sat, 10 Oct 2020 17:48:26 +0800
Subject: [PATCH] fix bugs

---
 mindspore/nn/layer/normalization.py  | 5 +++--
 mindspore/ops/_op_impl/tbe/neg.py    | 1 +
 mindspore/ops/operations/math_ops.py | 2 ++
 mindspore/ops/operations/nn_ops.py   | 6 +++---
 4 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/mindspore/nn/layer/normalization.py b/mindspore/nn/layer/normalization.py
index f340c959ec..1dd454b270 100644
--- a/mindspore/nn/layer/normalization.py
+++ b/mindspore/nn/layer/normalization.py
@@ -490,7 +490,7 @@ class LayerNorm(Cell):
     Args:
         normalized_shape (Union(tuple[int], list[int]): The normalization is performed over axis
             `begin_norm_axis ... R - 1`.
-        begin_norm_axis (int): It first normalization dimension: normalization will be performed along dimensions
+        begin_norm_axis (int): The first normalization dimension: normalization will be performed along dimensions
             `begin_norm_axis: rank(inputs)`, the value should be in [-1, rank(input)). Default: -1.
         begin_params_axis (int): The first parameter(beta, gamma)dimension: scale and centering parameters
             will have dimensions `begin_params_axis: rank(inputs)` and will be broadcast with
@@ -514,7 +514,8 @@ class LayerNorm(Cell):
         >>> x = Tensor(np.ones([20, 5, 10, 10]), mindspore.float32)
         >>> shape1 = x.shape[1:]
         >>> m = nn.LayerNorm(shape1,  begin_norm_axis=1, begin_params_axis=1)
-        >>> m(x)
+        >>> m(x).shape
+        (20, 5, 10, 10)
     """
 
     def __init__(self,
diff --git a/mindspore/ops/_op_impl/tbe/neg.py b/mindspore/ops/_op_impl/tbe/neg.py
index ee281a3bb2..5b44ab4699 100644
--- a/mindspore/ops/_op_impl/tbe/neg.py
+++ b/mindspore/ops/_op_impl/tbe/neg.py
@@ -29,6 +29,7 @@ neg_op_info = TBERegOp("Neg") \
     .dtype_format(DataType.I32_None, DataType.I32_None) \
     .dtype_format(DataType.F16_None, DataType.F16_None) \
     .dtype_format(DataType.F32_None, DataType.F32_None) \
+    .dtype_format(DataType.I8_None, DataType.I8_None) \
     .get_op_info()
 
 
diff --git a/mindspore/ops/operations/math_ops.py b/mindspore/ops/operations/math_ops.py
index 7ec6a39cad..3b26079fdb 100644
--- a/mindspore/ops/operations/math_ops.py
+++ b/mindspore/ops/operations/math_ops.py
@@ -449,6 +449,8 @@ class ReduceAny(_Reduce):
         >>> input_x = Tensor(np.array([[True, False], [True, True]]))
         >>> op = P.ReduceAny(keep_dims=True)
         >>> output = op(input_x, 1)
+        [[True],
+         [True]]
     """
 
     def __infer__(self, input_x, axis):
diff --git a/mindspore/ops/operations/nn_ops.py b/mindspore/ops/operations/nn_ops.py
index 934f72d209..24857d08bd 100644
--- a/mindspore/ops/operations/nn_ops.py
+++ b/mindspore/ops/operations/nn_ops.py
@@ -5549,10 +5549,10 @@ class DynamicRNN(PrimitiveWithInfer):
           The data type must be float16 or float32.
         - **b** (Tensor) - Bias. Tensor of shape (`4 x hidden_size`).
           The data type must be float16 or float32.
-        - **seq_length (Tensor) - The length of each batch. Tensor of shape (`batch_size`).
+        - **seq_length** (Tensor) - The length of each batch. Tensor of shape (`batch_size`).
           Only `None` is currently supported.
-        - **init_h (Tensor) - Hidden state of initial time. Tensor of shape (1, `batch_size`, `hidden_size`).
-        - **init_c (Tensor) - Cell state of initial time. Tensor of shape (1, `batch_size`, `hidden_size`).
+        - **init_h** (Tensor) - Hidden state of initial time. Tensor of shape (1, `batch_size`, `hidden_size`).
+        - **init_c** (Tensor) - Cell state of initial time. Tensor of shape (1, `batch_size`, `hidden_size`).
 
     Outputs:
         - **y** (Tensor) - A Tensor of shape (`num_step`, `batch_size`, `hidden_size`).