fix softmax

mindspore/ccsrc/backend/kernel_compiler/cpu/mkldnn/softmax_cpu_kernel.cc

@@ -28,9 +28,12 @@ void SoftmaxCPUKernel::InitKernel(const CNodePtr &kernel_node) {
     MS_LOG(EXCEPTION) << "cpu softmax only support input axis size 1";
   }
   int axis = axis_list[0];
-  if (axis == -1 || axis >= SizeToInt(src_shape.size())) {
+  if (axis >= SizeToInt(src_shape.size())) {
     axis = SizeToInt(src_shape.size()) - 1;
   }
+  while (axis < 0) {
+    axis += SizeToInt(src_shape.size());
+  }
   dnnl::memory::desc src_desc = GetDefaultMemDesc(src_shape);
   dnnl::softmax_forward::desc desc = dnnl::softmax_forward::desc(dnnl::prop_kind::forward_training, src_desc, axis);
   auto prim_desc = dnnl::softmax_forward::primitive_desc(desc, MKLKernelEngine::Get().engine());

tests/st/ops/cpu/test_softmax_op.py

@@ -29,7 +29,7 @@ context.set_context(mode=context.GRAPH_MODE, device_target='CPU')
 class NetSoftmax(nn.Cell):
     def __init__(self):
         super(NetSoftmax, self).__init__()
-        self.softmax = P.Softmax()
+        self.softmax = P.Softmax(axis=-1)
         x = Tensor(np.array([[0.1, 0.3, 0.6],
                              [0.2, -0.6, 0.8],
                              [0.6, 1, 0.4]]).astype(np.float32))
@@ -52,3 +52,31 @@ def test_softmax():
     diff = np.abs(outputSum - expect)
     print(diff)
     assert np.all(diff < error)
+
+
+class NetSoftmax1(nn.Cell):
+    def __init__(self):
+        super(NetSoftmax1, self).__init__()
+        self.softmax = P.Softmax(axis=-2)
+        x = Tensor(np.array([[0.1, 0.3, 0.6],
+                             [0.2, -0.6, 0.8],
+                             [0.6, 1, 0.4]]).astype(np.float32))
+        self.x = Parameter(initializer(x, x.shape), name='x')
+
+    def construct(self):
+        return self.softmax(self.x)
+
+
+@pytest.mark.level0
+@pytest.mark.platform_x86_cpu
+@pytest.mark.env_onecard
+def test_softmax1():
+    Softmax = NetSoftmax1()
+    output = Softmax()
+    output = output.asnumpy()
+    outputSum = output.sum(axis=0)
+    expect = np.ones(3)
+    error = expect * 1.0e-6
+    diff = np.abs(outputSum - expect)
+    print(diff)
+    assert np.all(diff < error)