!437 Add L2Loss op for VM.

Merge pull request !437 from liuxiao/add-l2-loss

mindspore/ops/_grad/grad_nn_ops.py

@@ -456,6 +456,17 @@ def get_bprop_smooth_l1_loss(self):
     return bprop
 
 
+@bprop_getters.register(P.L2Loss)
+def get_bprop_l2_loss(self):
+    """Grad definition for `L2Loss` operation."""
+
+    def bprop(x, out, dout):
+        dx = x * dout
+        return (dx,)
+
+    return bprop
+
+
 @bprop_getters.register(P.PReLU)
 def get_bprop_prelu(self):
     """Grad definition for `PReLU` operation."""

mindspore/ops/_op_impl/tbe/__init__.py

@@ -117,6 +117,7 @@ from .layer_norm_beta_gamma_backprop import _layer_norm_beta_gamma_backprop_tbe
 from .layer_norm import _layer_norm_tbe
 from .layer_norm_grad import _layer_norm_grad_tbe
 from .layer_norm_x_backprop import _layer_norm_x_backprop_tbe
+from .l2_loss import _l2_loss_tbe
 from .square_sum_v1 import _square_sum_v1_tbe
 from .square_sum_v2 import _square_sum_v2_tbe
 from .confusion_transpose_d import _confusion_transpose_d_tbe

mindspore/ops/_op_impl/tbe/l2_loss.py

+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ============================================================================
+
+"""L2Loss op"""
+from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
+
+l2_loss_op_info = TBERegOp("L2Loss") \
+    .fusion_type("OPAQUE") \
+    .async_flag(False) \
+    .binfile_name("l2_loss.so") \
+    .compute_cost(10) \
+    .kernel_name("l2_loss") \
+    .partial_flag(True) \
+    .input(0, "x", None, "required", None) \
+    .output(0, "y", True, "required", "all") \
+    .dtype_format(DataType.F16_Default, DataType.F16_Default) \
+    .dtype_format(DataType.F16_FracZ, DataType.F16_Default) \
+    .dtype_format(DataType.F16_FracNZ, DataType.F16_Default) \
+    .dtype_format(DataType.F16_5HD, DataType.F16_Default) \
+    .dtype_format(DataType.F16_C1HWNCoC0, DataType.F16_Default) \
+    .dtype_format(DataType.F32_Default, DataType.F32_Default) \
+    .dtype_format(DataType.F32_FracZ, DataType.F32_Default) \
+    .dtype_format(DataType.F32_FracNZ, DataType.F32_Default) \
+    .dtype_format(DataType.F32_5HD, DataType.F32_Default) \
+    .dtype_format(DataType.F32_C1HWNCoC0, DataType.F32_Default) \
+    .get_op_info()
+
+
+@op_info_register(l2_loss_op_info)
+def _l2_loss_tbe():
+    """L2Loss TBE register"""
+    return

mindspore/ops/operations/__init__.py

@@ -55,7 +55,7 @@ from .nn_ops import (LSTM, SGD, Adam, ApplyMomentum, BatchNorm,
                      DropoutDoMask,
                      DropoutGenMask, Flatten, FusedBatchNorm,
                      Gelu, Elu,
-                     GetNext, L2Normalize, LayerNorm,
+                     GetNext, L2Normalize, LayerNorm, L2Loss,
                      LogSoftmax,
                      MaxPool, ExtractImagePatches,
                      AvgPool, Conv2DBackpropInput,
@@ -167,6 +167,7 @@ __all__ = [
     'FloatStatus',
     'Reciprocal',
     'SmoothL1Loss',
+    'L2Loss',
     'ReduceAll',
     'ScalarToArray',
     'ScalarToTensor',

mindspore/ops/operations/nn_ops.py

@@ -1332,6 +1332,41 @@ class SmoothL1Loss(PrimitiveWithInfer):
         return prediction
 
 
+class L2Loss(PrimitiveWithInfer):
+    """
+    Calculates half of the L2 norm of a tensor without using the `sqrt`.
+
+    Set `input_x` as x and output as loss.
+
+    .. math::
+        loss = sum(x ** 2) / 2
+
+    Inputs:
+        - **input_x** (Tensor) - A input Tensor.
+
+    Outputs:
+        Tensor. Has the same dtype as `input_x`. The output tensor is the value of loss which is a scalar tensor.
+
+    Examples
+        >>> input_x = Tensor(np.array([1, 2, 3]), mindspore.float16)
+        >>> l2_loss = P.L2Loss()
+        >>> l2_loss(input_x)
+        7.0
+    """
+    @prim_attr_register
+    def __init__(self):
+        """init L2Loss"""
+
+    def infer_shape(self, input_x):
+        loss_shape = []
+        return loss_shape
+
+    def infer_dtype(self, x_type):
+        validator.check_subclass("x_type", x_type, mstype.tensor, self.name)
+        validator.check_tensor_type_same({'x_type': x_type}, [mstype.double, mstype.float_, mstype.float16], self.name)
+        return x_type
+
+
 class SGD(PrimitiveWithInfer):
     """
     Computes stochastic gradient descent (optionally with momentum).

tests/ut/python/ops/test_ops.py

@@ -871,6 +871,14 @@ test_case_nn_ops = [
         'desc_inputs': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3]],
         'desc_bprop': [3, 3],
         'skip': ['backward']}),
+    ('L2Loss_1', {
+        'block': P.L2Loss(),
+        'desc_inputs': [Tensor(np.array([1, 2, 3, 4]), mstype.float16)],
+        'desc_bprop': []}),
+    ('L2Loss_2', {
+        'block': P.L2Loss(),
+        'desc_inputs': [Tensor(np.array([[1, 1], [2, 2], [3, 3], [4, 4]]), mstype.float16)],
+        'desc_bprop': []}),
 ]
 
 test_case_array_ops = [