问题现象描述

系统环境：

Hardware Environment(Ascend/GPU/CPU): Ascend Software Environment: -- MindSpore version (source or binary): 1.6.0 -- Python version (e.g., Python 3.7.5): 3.7.6 -- OS platform and distribution (e.g., Linux Ubuntu 16.04): Ubuntu 4.15.0-74-generic -- GCC/Compiler version (if compiled from source):

训练脚本是通过构建ReduceSum的单算子网络，通过对维度中的所有元素求和来减少张量的维度。脚本如下：

class Net(Cell):
  def __init__(self, axis, keep_dims):
   super().__init__()
   self.reducesum = ops.ReduceSum(keep_dims=keep_dims)
   self.axis = axis
  def construct(self, input_x):
   return self.reducesum(input_x, self.axis)

x = Tensor(np.random.randn(10, 5, 4, 4, 4, 4, 4, 4, 4, 4),mindspore.float32)
net = Net(axis=(1,), keep_dims=True)
out = net(x)
print("out",out.shape)
#print(out)

报错信息如下：

Traceback (most recent call last):
File "demo4.py", line 11, in <module>
out = net(x)
…
RuntimeError: ({'errCode': 'E80012', 'op_name': 'reduce_sum_d', 'param_name': 'x', 'min_value': 0, 'max_value': 8, 'real_value': 10}, 'In op, the num of dimensions of input/output[x] should be inthe range of [0, 8], but actually is [10].')
The function call stack:
In file demo4.py(07)/    return self.reducesum(input_x, self.axis)/

原因分析

在RuntimeError中，max_value为8，而real_value为10，超过了ReduceSum算子支持的维度，提示输入参数维度应该在[0,8]之间，在官网中对ReduceSum也做了输入维度限制说明。

检查代码发现，09行代码维度为10，变量定义存在问题。需要将值设定为[0,8]。

参考文档：ReduceSum算子API接口

解决措施

基于以上原因，修改训练脚本：

class Net(Cell):
  def __init__(self, axis, keep_dims):
   super().__init__()
   self.reducesum = ops.ReduceSum(keep_dims=keep_dims)
   self.axis = axis
  def construct(self, input_x):
   return self.reducesum(input_x, self.axis)

x = Tensor(np.random.randn(10, 5, 4, 4, 4, 4, 4, 4),mindspore.float32)  # 将输入修改为8维
net = Net(axis=(1,), keep_dims=True)
out = net(x)
print("out",out.shape)
#print(out)

执行训练。成功后输出如下：

 out (10, 1, 4, 4, 4, 4, 4, 4)