Fix default value of fluid.memory_optimize (#18295)

* fix default value of fluid.memory_optimize
test=develop

* fix api.spec
test=develop

paddle/fluid/API.spec

@@ -27,7 +27,7 @@ paddle.fluid.DistributeTranspiler.get_pserver_programs (ArgSpec(args=['self', 'e
 paddle.fluid.DistributeTranspiler.get_startup_program (ArgSpec(args=['self', 'endpoint', 'pserver_program', 'startup_program'], varargs=None, keywords=None, defaults=(None, None)), ('document', '90a40b80e0106f69262cc08b861c3e39'))
 paddle.fluid.DistributeTranspiler.get_trainer_program (ArgSpec(args=['self', 'wait_port'], varargs=None, keywords=None, defaults=(True,)), ('document', '0e47f020304e2b824e87ff03475c17cd'))
 paddle.fluid.DistributeTranspiler.transpile (ArgSpec(args=['self', 'trainer_id', 'program', 'pservers', 'trainers', 'sync_mode', 'startup_program', 'current_endpoint'], varargs=None, keywords=None, defaults=(None, '127.0.0.1:6174', 1, True, None, '127.0.0.1:6174')), ('document', '418c7e8b268e9be4104f2809e654c2f7'))
-paddle.fluid.memory_optimize (ArgSpec(args=['input_program', 'skip_opt_set', 'print_log', 'level', 'skip_grads'], varargs=None, keywords=None, defaults=(None, False, 0, False)), ('document', '2348247f684bfd5bb9466470f35be064'))
+paddle.fluid.memory_optimize (ArgSpec(args=['input_program', 'skip_opt_set', 'print_log', 'level', 'skip_grads'], varargs=None, keywords=None, defaults=(None, False, 0, True)), ('document', '2348247f684bfd5bb9466470f35be064'))
 paddle.fluid.release_memory (ArgSpec(args=['input_program', 'skip_opt_set'], varargs=None, keywords=None, defaults=(None,)), ('document', 'd38c5b8b2b2e0bb19bcf1b581a80a7e4'))
 paddle.fluid.DistributeTranspilerConfig.__init__ 
 paddle.fluid.ParallelExecutor.__init__ (ArgSpec(args=['self', 'use_cuda', 'loss_name', 'main_program', 'share_vars_from', 'exec_strategy', 'build_strategy', 'num_trainers', 'trainer_id', 'scope'], varargs=None, keywords=None, defaults=(None, None, None, None, None, 1, 0, None)), ('document', '6adf97f83acf6453d4a6a4b1070f3754'))
@@ -729,7 +729,7 @@ paddle.fluid.transpiler.DistributeTranspiler.get_pserver_programs (ArgSpec(args=
 paddle.fluid.transpiler.DistributeTranspiler.get_startup_program (ArgSpec(args=['self', 'endpoint', 'pserver_program', 'startup_program'], varargs=None, keywords=None, defaults=(None, None)), ('document', '90a40b80e0106f69262cc08b861c3e39'))
 paddle.fluid.transpiler.DistributeTranspiler.get_trainer_program (ArgSpec(args=['self', 'wait_port'], varargs=None, keywords=None, defaults=(True,)), ('document', '0e47f020304e2b824e87ff03475c17cd'))
 paddle.fluid.transpiler.DistributeTranspiler.transpile (ArgSpec(args=['self', 'trainer_id', 'program', 'pservers', 'trainers', 'sync_mode', 'startup_program', 'current_endpoint'], varargs=None, keywords=None, defaults=(None, '127.0.0.1:6174', 1, True, None, '127.0.0.1:6174')), ('document', '418c7e8b268e9be4104f2809e654c2f7'))
-paddle.fluid.transpiler.memory_optimize (ArgSpec(args=['input_program', 'skip_opt_set', 'print_log', 'level', 'skip_grads'], varargs=None, keywords=None, defaults=(None, False, 0, False)), ('document', '2348247f684bfd5bb9466470f35be064'))
+paddle.fluid.transpiler.memory_optimize (ArgSpec(args=['input_program', 'skip_opt_set', 'print_log', 'level', 'skip_grads'], varargs=None, keywords=None, defaults=(None, False, 0, True)), ('document', '2348247f684bfd5bb9466470f35be064'))
 paddle.fluid.transpiler.release_memory (ArgSpec(args=['input_program', 'skip_opt_set'], varargs=None, keywords=None, defaults=(None,)), ('document', 'd38c5b8b2b2e0bb19bcf1b581a80a7e4'))
 paddle.fluid.transpiler.HashName.__init__ (ArgSpec(args=['self', 'pserver_endpoints'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754'))
 paddle.fluid.transpiler.HashName.dispatch (ArgSpec(args=['self', 'varlist'], varargs=None, keywords=None, defaults=None), ('document', '6adf97f83acf6453d4a6a4b1070f3754'))

python/paddle/fluid/transpiler/memory_optimization_transpiler.py

@@ -497,7 +497,7 @@ def memory_optimize(input_program,
                     skip_opt_set=None,
                     print_log=False,
                     level=0,
-                    skip_grads=False):
+                    skip_grads=True):
     """
     | Legacy memory optimization strategy, reduce total memory consumption by reuse variable memory between different operators.
     | Simple sample to explain the algorithm: