:[gdZddlZddlZddlZddlZddlmZmZddlm Z m Z m Z m Z ddl ZddlmZddlmZmZmZmZmZmZej0dZd e d dfd Zdd Zd efd Zd efdZd efdZde d dfdZ d e!fdZ"deded efdZ#dejHd efdZ%eGddeZ&dejNde&d ejNfdZ(ddZ)GddZ*y) z-XGBoost collective communication related API.N)IntEnumunique)AnyDictListOptional)_T)_LIB _check_call build_infoc_str make_jcargspy_strz[xgboost.collective]argsreturnc Rttjtdi|y)aInitialize the collective library with arguments. Parameters ---------- args : Keyword arguments representing the parameters and their values. Accepted parameters: - dmlc_communicator: The type of the communicator. * rabit: Use Rabit. This is the default if the type is unspecified. * federated: Use the gRPC interface for Federated Learning. Only applicable to the Rabit communicator: - dmlc_tracker_uri: Hostname of the tracker. - dmlc_tracker_port: Port number of the tracker. - dmlc_task_id: ID of the current task, can be used to obtain deterministic - dmlc_retry: The number of retry when handling network errors. - dmlc_timeout: Timeout in seconds. - dmlc_nccl_path: Path to load (dlopen) nccl for GPU-based communication. Only applicable to the Federated communicator (use upper case for environment variables, use lower case for runtime configuration): - federated_server_address: Address of the federated server. - federated_world_size: Number of federated workers. - federated_rank: Rank of the current worker. - federated_server_cert: Server certificate file path. Only needed for the SSL mode. - federated_client_key: Client key file path. Only needed for the SSL mode. - federated_client_cert: Client certificate file path. Only needed for the SSL mode. N)r r XGCommunicatorInitr)rs M/var/www/html/bid-api/venv/lib/python3.12/site-packages/xgboost/collective.pyinitrs B'' (;d(;<=c>ttjy)zFinalize the communicator.N)r r XGCommunicatorFinalizerrrfinalizer6s++-.rc.tj}|S)zjGet rank of current process. Returns ------- rank : int Rank of current process. )r XGCommunicatorGetRankrets rget_rankr ;s  $ $ &C Jrc.tj}|S)zdGet total number workers. Returns ------- n : int Total number of process. )r XGCommunicatorGetWorldSizers rget_world_sizer#Gs  ) ) +C Jrc.tj}|S)z.If the collective communicator is distributed.)r XGCommunicatorIsDistributed)is_dists ris_distributedr'Ss..0G Nrmsgct|ts t|}tj}|dk7r6t tj t |jyt|jdy)zPrint message to the communicator. This function can be used to communicate the information of the progress to the communicator. Parameters ---------- msg : str The message to be printed to the communicator. rT)flushN) isinstancestrr r%r XGCommunicatorPrintrstripprint)r(r&s rcommunicator_printr0YsZ c3 #h..0G!|D,,U399;-?@A ciik&rctj}ttjtj ||j }|sJt|S)zhGet the processor name. Returns ------- name : str the name of processor(host) )ctypesc_char_pr r XGCommunicatorGetProcessorNamebyrefvaluer)name_strr6s rget_processor_namer8msF H33FLL4JKL NNE L5 %=rdatarootc Xt}tj}||k(r>|Jdtj|tj }t ||_ttjtj|tjtj|||k7rtj|jz}ttjtj|tj|j|tj |j"}~|Sttjtjtj$tj|j|~|S)aSBroadcast object from one node to all other nodes. Parameters ---------- data : any type that can be pickled Input data, if current rank does not equal root, this can be None root : int Rank of the node to broadcast data from. Returns ------- object : int the result of broadcast. z&need to pass in data when broadcasting)protocol)r r2c_ulongpickledumpsHIGHEST_PROTOCOLlenr6r r XGCommunicatorBroadcastr5sizeofc_charcastc_void_ploadsrawr3)r9r:ranklengthsdptrs r broadcastrM|s6 :D ^^ F t|I!II LL(?(? @1v  $$ LL &--"?   t|  ,/  ( ( D&//2FLL$  ||DHH%  K   ( ( FOOA.@&,,PT   Krdtypecptjddtjddtjddtjddtjd d tjd d tjd dtjddtjddtjddtjddi } |jtjddi||vrtd|d||S#t$rY#wxYw)Nfloat16rfloat32r float64int8int16int32int64uint8uint16 uint32 uint64 float128z data type z* is not supported on the current platform.)nprNupdate TypeError)rN dtype_maps r _map_dtyperjs Q Q Q ! 1 1 1 1 A B B I "((:.23 I*UG+UVWW U     s+&D)) D54D5c(eZdZdZdZdZdZdZdZdZ y) Opz#Supported operations for allreduce.rr rSrerUrWN) __name__ __module__ __qualname____doc__MAXMINSUM BITWISE_AND BITWISE_OR BITWISE_XORrrrrlrls#- C C CKJKrrlopc ht|tjs td|j j }t tj|jjtj|jt|jt||S)a'Perform allreduce, return the result. Parameters ---------- data : Input data. op : Reduction operator. Returns ------- result : The result of allreduce, have same shape as data Notes ----- This function is not thread-safe. z%allreduce only takes in numpy.ndarray)r+rfndarrayrhravelcopyr r XGCommunicatorAllreducer2data_asrFsizerjrNint)r9rwbufs r allreducers& dBJJ '?@@ **,   C $$ JJ  v / HH syy ! G   Jrc>ttjy)zKill the process.N)r r XGCommunicatorSignalErrorrrr signal_errorrs..01rcFeZdZdZdeddfdZdeeeffdZde ddfdZ y)CommunicatorContextzNA context controlling collective communicator initialization and finalization.rrNc \||_d}|j|dyt}|dsy ddlm}|j *t jj|j }nd}|r0t jj|d}||j|<yy#t$rYywxYw)Ndmlc_nccl_pathUSE_DLOPEN_NCCLr)libz libnccl.so.2) rgetr nvidia.ncclr__file__ospathdirnamejoin ImportError)selfrkeybinforrrs r__init__zCommunicatorContext.__init__s  88C  *  &'   ' ||')+)Fww||G^<!% #   sA/B B+*B+ctdi|jtsJtj d|jS)Nz8-------------- communicator say hello ------------------r)rrr'LOGGERdebug)rs r __enter__zCommunicatorContext.__enter__s4 tyy OPyyrcBttjdy)Nz7--------------- communicator say bye ------------------)rrr)rrs r__exit__zCommunicatorContext.__exit__s  NOr) rmrnrorprrrr,rrrrrrrrsBXst84S> PdPtPrr)rN)+rpr2loggingrr>enumrrtypingrrrrnumpyrf_typingr corer r r rrr getLoggerrrrrr r#r'r0r,r8rMrNrjrlryrrrrrrrs3  ,,KK   1 2!>!>!>H/ #    'C'D'( C ,B,c,b,`bhh32BJJB2::B2 'P'Pr