Merge pull request #15902 from colourful-tree/new_develop

remove mkldnn & fix commit

paddle/fluid/API.spec

@@ -92,7 +92,7 @@ paddle.fluid.layers.pool3d ArgSpec(args=['input', 'pool_size', 'pool_type', 'poo
 paddle.fluid.layers.adaptive_pool2d ArgSpec(args=['input', 'pool_size', 'pool_type', 'require_index', 'name'], varargs=None, keywords=None, defaults=('max', False, None))
 paddle.fluid.layers.adaptive_pool3d ArgSpec(args=['input', 'pool_size', 'pool_type', 'require_index', 'name'], varargs=None, keywords=None, defaults=('max', False, None))
 paddle.fluid.layers.batch_norm ArgSpec(args=['input', 'act', 'is_test', 'momentum', 'epsilon', 'param_attr', 'bias_attr', 'data_layout', 'in_place', 'name', 'moving_mean_name', 'moving_variance_name', 'do_model_average_for_mean_and_var', 'fuse_with_relu', 'use_global_stats'], varargs=None, keywords=None, defaults=(None, False, 0.9, 1e-05, None, None, 'NCHW', False, None, None, None, False, False, False))
-paddle.fluid.layers.data_norm ArgSpec(args=['input', 'act', 'epsilon', 'param_attr', 'data_layout', 'in_place', 'use_mkldnn', 'name', 'moving_mean_name', 'moving_variance_name', 'do_model_average_for_mean_and_var'], varargs=None, keywords=None, defaults=(None, 1e-05, None, 'NCHW', False, False, None, None, None, False))
+paddle.fluid.layers.data_norm ArgSpec(args=['input', 'act', 'epsilon', 'param_attr', 'data_layout', 'in_place', 'name', 'moving_mean_name', 'moving_variance_name', 'do_model_average_for_mean_and_var'], varargs=None, keywords=None, defaults=(None, 1e-05, None, 'NCHW', False, None, None, None, False))
 paddle.fluid.layers.beam_search_decode ArgSpec(args=['ids', 'scores', 'beam_size', 'end_id', 'name'], varargs=None, keywords=None, defaults=(None,))
 paddle.fluid.layers.conv2d_transpose ArgSpec(args=['input', 'num_filters', 'output_size', 'filter_size', 'padding', 'stride', 'dilation', 'groups', 'param_attr', 'bias_attr', 'use_cudnn', 'act', 'name'], varargs=None, keywords=None, defaults=(None, None, 0, 1, 1, None, None, None, True, None, None))
 paddle.fluid.layers.conv3d_transpose ArgSpec(args=['input', 'num_filters', 'output_size', 'filter_size', 'padding', 'stride', 'dilation', 'groups', 'param_attr', 'bias_attr', 'use_cudnn', 'act', 'name'], varargs=None, keywords=None, defaults=(None, None, 0, 1, 1, None, None, None, True, None, None))

paddle/fluid/operators/data_norm_op.cc

@@ -140,9 +140,6 @@ class DataNormOpMaker : public framework::OpProtoAndCheckerMaker {
               "Scales of the history data batch, "
               "will apply to output when training")
         .AsIntermediate();
-    AddAttr<bool>("use_mkldnn",
-                  "(bool, default false) Only used in mkldnn kernel")
-        .SetDefault(false);
     AddComment(R"DOC(
 Data Normalization.
 

paddle/fluid/operators/teacher_student_sigmoid_loss_op.cc

@@ -117,11 +117,11 @@ class TeacherStudentSigmoidLossOpMaker
               "[N x 1]. The teacher student sigmoid loss.");
     AddAttr<float>(
         "soft_max_up_bound",
-        "fp32, if input > soft_max_up_bound, will be bound, default 15.0")
+        "fp32, if input > soft_max_up_bound, input will be bound, default 15.0")
         .SetDefault(15.0);
-    AddAttr<float>(
+    AddAttr<float>("soft_max_lower_bound",
-        "soft_max_lower_bound",
+                   "fp32, if input < soft_max_lower_bound, input will be "
-        "fp32, if input < soft_max_lower_bound, will be bound, default -15.0")
+                   "bound, default -15.0")
         .SetDefault(-15.0);
     AddComment(R"DOC(
 TeacherStudentSigmoidLoss Operator.
@@ -134,7 +134,7 @@ we add another label(z') to original.
         label = {-2, -1, [0, 2]}
         when z' is not exist, clk = 0 : label = -2;
         when z' is not exist, clk = 1 : label = -1;
-        when z' is exist    , clk = 0 : label = 0 + z';
+        when z' is exist , clk = 0 : label = 0 + z';
         when z' is exist    , clk = 1 : label = 1 + z';
 
 )DOC");

python/paddle/fluid/layers/nn.py

@@ -3041,7 +3041,6 @@ def data_norm(input,
               param_attr=None,
               data_layout='NCHW',
               in_place=False,
-              use_mkldnn=False,
               name=None,
               moving_mean_name=None,
               moving_variance_name=None,
@@ -3075,7 +3074,6 @@ def data_norm(input,
         param_attr(ParamAttr): The parameter attribute for Parameter `scale`.
         data_layout(string, default NCHW): NCHW|NHWC
         in_place(bool, Default False): Make the input and output of batch norm reuse memory.
-        use_mkldnn(bool, Default false): ${use_mkldnn_comment}
         name(string, Default None): A name for this layer(optional). If set None, the layer
             will be named automatically.
         moving_mean_name(string, Default None): The name of moving_mean which store the global Mean.
@@ -3156,8 +3154,7 @@ def data_norm(input,
         outputs={"Y": data_norm_out,
                  "Means": means,
                  "Scales": scales},
-        attrs={"epsilon": epsilon,
+        attrs={"epsilon": epsilon})
-               "use_mkldnn": use_mkldnn})
 
     return helper.append_activation(data_norm_out)
 
@@ -9945,6 +9942,7 @@ def teacher_student_sigmoid_loss(input,
 
     Examples:
         .. code-block:: python
+
           cost = fluid.layers.teacher_student_sigmoid_loss(input=similarity, label=label)
     """
     helper = LayerHelper('teacher_student_sigmoid_loss', **locals())