Created by: qianledan

memory_optimize is deprecated. Use CompiledProgram and Executor ParallelExecutor is deprecated. Please use CompiledProgram and Executor. CompiledProgram is a central place for optimization and Executor is the unified executor. Example can be found in compiler.py. W0421 19:46:12.907804 10233 graph.h:204] WARN: After a series of passes, the current graph can be quite different from OriginProgram. So, please avoid using the OriginProgram() method! W0421 19:46:13.659762 10233 device_context.cc:261] Please NOTE: device: 0, CUDA Capability: 37, Driver API Version: 9.0, Runtime API Version: 9.0 W0421 19:46:13.663028 10233 device_context.cc:269] device: 0, cuDNN Version: 7.0. I0421 19:46:15.136268 10233 build_strategy.cc:282] SeqOnlyAllReduceOps:0, num_trainers:1 ParallelExecutor is deprecated. Please use CompiledProgram and Executor. CompiledProgram is a central place for optimization and Executor is the unified executor. Example can be found in compiler.py. W0421 19:46:15.188122 10233 graph.h:204] WARN: After a series of passes, the current graph can be quite different from OriginProgram. So, please avoid using the OriginProgram() method! share_vars_from is set, scope is ignored. I0421 19:46:15.251974 10233 build_strategy.cc:282] SeqOnlyAllReduceOps:0, num_trainers:1