一import paddle 就打印一堆信息

[zzzzzz0407]

如图，也不影响使用，但很奇怪期望可以被解决
In [1]: import paddle
WARNING: Logging before InitGoogleLogging() is written to STDERR
I0115 16:42:38.097568 1452196 dynamic_loader.cc:194] Set paddle lib path : /home/zhangrufeng/anaconda3/envs/LuckyPaddle/lib/python3.8/site-packages/paddle/libs
I0115 16:42:38.219532 1452196 init.cc:96] Before Parse: argc is 2, Init commandline: dummy --tryfromenv=use_mkldnn,enable_exit_when_partial_worker,einsum_opt,eager_delete_tensor_gb,graph_load_in_parallel,allocator_strategy,enable_all2all_use_fp16,cublaslt_exhaustive_search_times,gpu_allocator_retry_time,use_cuda_managed_memory,check_kernel_launch,static_executor_perfstat_filepath,cudnn_exhaustive_search,gpu_memory_limit_mb,add_dependency_for_communication_op,new_executor_use_cuda_graph,reader_queue_speed_test_mode,gpugraph_dedup_pull_push_mode,gpugraph_enable_segment_merge_grads,convert_all_blocks,low_precision_op_list,initial_cpu_memory_in_mb,eager_delete_scope,gpugraph_load_node_list_into_hbm,check_nan_inf,use_shm_cache,tracer_mkldnn_ops_off,embedding_deterministic,free_idle_chunk,gpugraph_enable_gpu_direct_access,gemm_use_half_precision_compute_type,cudnn_batchnorm_spatial_persistent,dygraph_debug,host_trace_level,gpugraph_storage_mode,fuse_parameter_groups_size,max_inplace_grad_add,use_autotune,communicator_max_merge_var_num,gpugraph_enable_hbm_table_collision_stat,conv_workspace_size_limit,new_executor_use_inplace,local_exe_sub_scope_limit,fuse_parameter_memory_size,selected_gpus,rpc_send_thread_num,tensor_operants_mode,graph_metapath_split_opt,init_allocated_mem,dist_threadpool_size,new_executor_static_build,graph_get_neighbor_id,enable_cublas_tensor_op_math,enable_gpu_memory_usage_log,multiple_of_cupti_buffer_size,reallocate_gpu_memory_in_mb,use_stream_safe_cuda_allocator,rocksdb_path,use_system_allocator,jit_engine_type,gpugraph_debug_gpu_memory,get_host_by_name_time,enable_opt_get_features,use_fast_math,gpugraph_hbm_table_load_factor,gpugraph_sparse_table_storage_mode,trt_ibuilder_cache,run_kp_kernel,executor_log_deps_every_microseconds,communicator_is_sgd_optimizer,enable_auto_detect_gpu_topo,tracer_profile_fname,search_cache_max_number,gpugraph_merge_grads_segment_size,paddle_num_threads,conv2d_disable_cudnn,fraction_of_cuda_pinned_memory_to_use,new_executor_sequential_run,allreduce_record_one_event,communicator_send_queue_size,check_nan_inf_level,enable_unused_var_check,enable_rpc_profiler,enable_auto_rdma_trans,fleet_executor_with_standalone,npu_storage_format,cpu_deterministic,free_when_no_cache_hit,fraction_of_cpu_memory_to_use,initial_gpu_memory_in_mb,cudnn_deterministic,apply_pass_to_program,sort_sum_gradient,use_pinned_memory,call_stack_level,cudnn_exhaustive_search_times,new_executor_use_local_scope,use_virtual_memory_auto_growth,pe_profile_fname,enable_tracker_all2all,nccl_blocking_wait,enable_sparse_inner_gather,new_executor_serial_run,memory_fraction_of_eager_deletion,enable_api_kernel_fallback,enable_gpu_memory_usage_log_mb,set_to_1d,print_sub_graph_dir,prim_enabled,inner_op_parallelism,tracer_mkldnn_ops_on,cache_inference_while_scope,sync_nccl_allreduce,new_executor_log_memory_stats,fast_eager_deletion_mode,gpugraph_slot_feasign_max_num,auto_growth_chunk_size_in_mb,fraction_of_gpu_memory_to_use,benchmark
I0115 16:42:38.219650 1452196 init.cc:104] After Parse: argc is 1

[warrentdrew]

您好，看起来应该是paddle框架环境配置相关的问题，建议在paddle框架repo中提出issue
https://github.com/PaddlePaddle/Paddle/issues