refine doc

paddleslim/dist/single_distiller.py

@@ -108,6 +108,7 @@ def fsp_loss(teacher_var1_name, teacher_var2_name, student_var1_name,
             second dimension, all other dimensions should
             be consistent with student_var1.
         program(Program): The input distiller program.
+                          default: fluid.default_main_program()
     Return(Variable): fsp distiller loss.
     """
     teacher_var1 = program.global_block().var(teacher_var1_name)
@@ -129,6 +130,7 @@ def l2_loss(teacher_var_name, student_var_name,
         teacher_var_name(str): The name of teacher_var.
         student_var_name(str): The name of student_var.
         program(Program): The input distiller program.
+                          default: fluid.default_main_program() 
     Return(Variable): l2 distiller loss.
     """
     student_var = program.global_block().var(student_var_name)
@@ -149,6 +151,7 @@ def soft_label_loss(teacher_var_name,
         teacher_var_name(str): The name of teacher_var.
         student_var_name(str): The name of student_var.
         program(Program): The input distiller program.
+                          default: fluid.default_main_program() 
         teacher_temperature(float): Temperature used to divide
             teacher_feature_map before softmax. default: 1.0
         student_temperature(float): Temperature used to divide 
@@ -171,6 +174,7 @@ def loss(loss_func, program=fluid.default_main_program(), **kwargs):
     Combine variables from student model and teacher model by self defined loss.
     Args:
         program(Program): The input distiller program.
+                          default: fluid.default_main_program() 
         loss_func(function): The user self defined loss function. 
     Return(Variable): self defined distiller loss.
     """