MCPcopy Create free account

hub / github.com/Rivendile/Muri / functions

Functions770 in github.com/Rivendile/Muri

⨍ Functions 770◇ Types & classes 115
↓ 1 callersMethodinit_log
(self)
cluster_exp/log.py:42
↓ 1 callersMethodinit_log
(self)
simulator/log.py:37
↓ 1 callersMethodinit_node
(self, num_gpu=0, num_cpu=0, mem=0)
cluster_exp/node.py:33
↓ 1 callersMethodinit_node
(self, num_gpu=0, num_cpu=0, mem=0)
simulator/node.py:33
↓ 1 callersMethodinit_stats
(self)
cluster_exp/trainer.py:41
↓ 1 callersFunctionis_plain_type
(x)
cluster_exp/workloads/models/deep_rl/utils/misc.py:65
↓ 1 callersMethodkill_workers
(self)
cluster_exp/controller.py:96
↓ 1 callersMethodload
(self, filename)
cluster_exp/workloads/models/deep_rl/agent/BaseAgent.py:30
↓ 1 callersFunctionload_and_cache_examples
(args, sargs, tokenizer)
cluster_exp/workloads/models/nlp_model.py:101
↓ 1 callersMethodload_results
(self, dirs, **kwargs)
cluster_exp/workloads/models/deep_rl/utils/plot.py:33
↓ 1 callersFunctionmain
()
cluster_exp/run.py:1703
↓ 1 callersFunctionmain
()
simulator/run_sim.py:2213
↓ 1 callersFunctionmake_env
(env_id, seed, rank, episode_life=True)
cluster_exp/workloads/models/deep_rl/component/envs.py:27
↓ 1 callersMethodmake_server_for_master
(self, port: int)
cluster_exp/worker.py:75
↓ 1 callersMethodmake_server_for_scheduler
(self, port: int)
cluster_exp/trainer.py:58
↓ 1 callersMethodmake_server_for_trainer
(self, port)
cluster_exp/scheduler.py:31
↓ 1 callersMethodmake_server_for_worker
(self, port: int)
cluster_exp/controller.py:35
↓ 1 callersFunctionmask_tokens
Prepare masked tokens inputs/labels for masked language modeling: 80% MASK, 10% random, 10% original.
cluster_exp/workloads/models/nlp_model.py:106
↓ 1 callersMethodmatching
(self)
cluster_exp/matching.py:441
↓ 1 callersMethodmatching
(self)
simulator/matching.py:499
↓ 1 callersMethodmin_load_nodes
return need_gpu gpus whose gpu util are minimum
cluster_exp/switch.py:514
↓ 1 callersMethodmin_load_nodes
return need_gpu gpus whose gpu util are minimum
simulator/switch.py:496
↓ 1 callersMethodmove_to_pending
(self, job)
cluster_exp/jobs.py:465
↓ 1 callersMethodmove_to_pending
(self, job)
simulator/jobs.py:473
↓ 1 callersMethodms_yarn_alloc_res
ms_yarn allocates res from a single switch, if no enough gpus, give up, return False (all-or-nothing) if need_gpu > gpu_p_n
cluster_exp/switch.py:221
↓ 1 callersMethodms_yarn_alloc_res
ms_yarn allocates res from a single switch, if no enough gpus, give up, return False (all-or-nothing) if need_gpu > gpu_p_n
simulator/switch.py:203
↓ 1 callersMethodnone_placement
(self, job)
cluster_exp/cluster.py:833
↓ 1 callersMethodnone_placement
(self, job)
simulator/cluster.py:831
↓ 1 callersFunctionparse_cluster_spec
()
cluster_exp/run.py:135
↓ 1 callersFunctionparse_cluster_spec
()
simulator/run_sim.py:139
↓ 1 callersFunctionparse_job_file
(trace_file)
cluster_exp/run.py:105
↓ 1 callersFunctionparse_job_file
(trace_file)
simulator/run_sim.py:109
↓ 1 callersMethodplot_median_std
(self, data, x=None, **kwargs)
cluster_exp/workloads/models/deep_rl/utils/plot.py:124
↓ 1 callersMethodprepare_job_start_events
add job start events into job_events list end events should be added when they are starting
cluster_exp/jobs.py:580
↓ 1 callersMethodprepare_job_start_events
add job start events into job_events list end events should be added when they are starting
simulator/jobs.py:588
↓ 1 callersMethodprint_cluster_spec
(self)
cluster_exp/cluster.py:69
↓ 1 callersMethodprint_cluster_spec
(self)
simulator/cluster.py:69
↓ 1 callersMethodprint_job_events
(self)
cluster_exp/jobs.py:553
↓ 1 callersMethodprint_job_events
(self)
simulator/jobs.py:561
↓ 1 callersMethodrandom_placement
randomly pick up enough resource for both PS and worker in a job allocate one by one
cluster_exp/cluster.py:454
↓ 1 callersMethodrandom_placement
randomly pick up enough resource for both PS and worker in a job allocate one by one
simulator/cluster.py:452
↓ 1 callersFunctionrandom_sample
(indices, batch_size)
cluster_exp/workloads/models/deep_rl/utils/misc.py:55
↓ 1 callersFunctionrandom_seed
(seed=None)
cluster_exp/workloads/models/deep_rl/utils/torch_utils.py:36
↓ 1 callersMethodreal_job
(self)
cluster_exp/task.py:37
↓ 1 callersMethodrecord
(self, iteration_time)
cluster_exp/trainer.py:50
↓ 1 callersMethodrecord_obs
(self, env, dir, steps)
cluster_exp/workloads/models/deep_rl/agent/BaseAgent.py:103
↓ 1 callersMethodrecord_step
(self, state)
cluster_exp/workloads/models/deep_rl/agent/BaseAgent.py:99
↓ 1 callersMethodregister
(self)
cluster_exp/trainer.py:32
↓ 1 callersMethodregister_trainer
(self, trainer_ip, trainer_port, job_id)
cluster_exp/runtime/rpc/trainer_client.py:29
↓ 1 callersMethodregister_worker
(self, worker_ip, worker_port, num_gpus)
cluster_exp/runtime/rpc/worker_client.py:29
↓ 1 callersMethodrelease_job_gpu
(self, num_job=1)
cluster_exp/jobs.py:63
↓ 1 callersMethodrelease_job_gpu
(self, num_job=1)
simulator/jobs.py:61
↓ 1 callersMethodrelease_network_load
(self, in_load=0, out_load=0)
cluster_exp/node.py:149
↓ 1 callersMethodrelease_network_load
(self, in_load=0, out_load=0)
simulator/node.py:147
↓ 1 callersMethodremove_from_pending
(self, job, event_time)
cluster_exp/jobs.py:457
↓ 1 callersMethodremove_from_pending
(self, job, event_time)
simulator/jobs.py:465
↓ 1 callersMethodreport_itertime
(self, job_id, iter_time, src_utils)
cluster_exp/runtime/rpc/trainer_client.py:40
↓ 1 callersMethodreserve_gpus
GPU cluster reserve gpus for gpu_job groups
cluster_exp/jobs.py:672
↓ 1 callersMethodreserve_gpus
GPU cluster reserve gpus for gpu_job groups
simulator/jobs.py:680
↓ 1 callersMethodreset
(self)
cluster_exp/workloads/models/deep_rl/component/replay.py:44
↓ 1 callersMethodreset_noise
(self)
cluster_exp/workloads/models/deep_rl/network/network_utils.py:73
↓ 1 callersMethodreset_parameters
(self)
cluster_exp/workloads/models/deep_rl/network/network_utils.py:64
↓ 1 callersMethodreset_states
(self)
cluster_exp/workloads/models/deep_rl/component/random_process.py:40
↓ 1 callersMethodrun_jobs_to_packing
(self, run_jobs)
cluster_exp/matching.py:635
↓ 1 callersMethodrun_jobs_to_packing
(self, run_jobs)
simulator/matching.py:693
↓ 1 callersMethodsample_option
(self, prediction, epsilon, prev_option, is_intial_states)
cluster_exp/workloads/models/deep_rl/agent/OptionCritic_agent.py:29
↓ 1 callersMethodsave
(self, filename)
cluster_exp/workloads/models/deep_rl/agent/BaseAgent.py:25
↓ 1 callersMethodsoft_update
(self, target, src)
cluster_exp/workloads/models/deep_rl/agent/DDPG_agent.py:26
↓ 1 callersMethodsoft_update
(self, target, src)
cluster_exp/workloads/models/deep_rl/agent/TD3_agent.py:26
↓ 1 callersMethodsort_all_jobs
Sort jobs based on their sumbit_time j1, num_gpu, start_t, end_t, duration
cluster_exp/jobs.py:393
↓ 1 callersMethodsort_all_jobs
Sort jobs based on their sumbit_time j1, num_gpu, start_t, end_t, duration
simulator/jobs.py:401
↓ 1 callersMethodstep
(self)
cluster_exp/workloads/models/deep_rl/agent/DQN_agent.py:101
↓ 1 callersFunctionswap
(x,y)
cluster_exp/matching.py:20
↓ 1 callersFunctionswap
(x,y)
simulator/matching.py:27
↓ 1 callersMethodswitch_task
(self)
cluster_exp/workloads/models/deep_rl/agent/BaseAgent.py:75
↓ 1 callersFunctionthemis_sim_jobs
new jobs are added to the end of the ending queue but in the queue, shortest (gpu) job first be served, until no resource
cluster_exp/run.py:1118
↓ 1 callersFunctionthemis_sim_jobs
themis finish-time fairness
simulator/run_sim.py:1823
↓ 1 callersFunctiontrain
()
cluster_exp/workloads/main_real_util.py:85
↓ 1 callersFunctiontrain
()
cluster_exp/workloads/main_real_preenv.py:83
↓ 1 callersMethodtry_cross_node_alloc_antman
used in MS_YARN placement try get gpus from multiple nodes [need_gpu / gpu_p_node] nodes, and one node with [need_gpu % g
cluster_exp/switch.py:247
↓ 1 callersMethodtry_cross_node_alloc_antman
used in MS_YARN placement try get gpus from multiple nodes [need_gpu / gpu_p_node] nodes, and one node with [need_gpu % g
simulator/switch.py:229
↓ 1 callersMethodtry_single_node_alloc_antman
used in MS_YARN placement try get gpus from a single node if can't find a node, give up, and return False
cluster_exp/switch.py:427
↓ 1 callersMethodtry_single_node_alloc_antman
used in MS_YARN placement try get gpus from a single node if can't find a node, give up, and return False
simulator/switch.py:409
↓ 1 callersMethodupdate_stats
(self, iteration_time)
cluster_exp/trainer.py:45
↓ 1 callersMethodvalid_index
(self, index)
cluster_exp/workloads/models/deep_rl/component/replay.py:105
↓ 1 callersMethodwait_for_workers
(self)
cluster_exp/controller.py:92
↓ 1 callersMethodzero
(self)
cluster_exp/workloads/models/deep_rl/utils/torch_utils.py:155
FunctionDEFINE_boolean
Defines a flag of type 'boolean'. Args: flag_name: The name of the flag as a string. default_value: The default value the flag should take
cluster_exp/flags.py:82
FunctionDEFINE_boolean
Defines a flag of type 'boolean'. Args: flag_name: The name of the flag as a string. default_value: The default value the flag should take
simulator/flags.py:82
FunctionDEFINE_float
Defines a flag of type 'float'. Args: flag_name: The name of the flag as a string. default_value: The default value the flag should take as
cluster_exp/flags.py:112
FunctionDEFINE_float
Defines a flag of type 'float'. Args: flag_name: The name of the flag as a string. default_value: The default value the flag should take as
simulator/flags.py:112
FunctionDEFINE_integer
Defines a flag of type 'int'. Args: flag_name: The name of the flag as a string. default_value: The default value the flag should take as a
cluster_exp/flags.py:71
FunctionDEFINE_integer
Defines a flag of type 'int'. Args: flag_name: The name of the flag as a string. default_value: The default value the flag should take as a
simulator/flags.py:71
FunctionDEFINE_string
Defines a flag of type 'string'. Args: flag_name: The name of the flag as a string. default_value: The default value the flag should take a
cluster_exp/flags.py:60
FunctionDEFINE_string
Defines a flag of type 'string'. Args: flag_name: The name of the flag as a string. default_value: The default value the flag should take a
simulator/flags.py:60
FunctionDEFINE_version
(v_string)
cluster_exp/flags.py:123
FunctionDEFINE_version
(v_string)
simulator/flags.py:123
MethodDone
Missing associated documentation comment in .proto file.
cluster_exp/runtime/rpc_stubs/worker_to_master_pb2_grpc.py:38
MethodDone
(request, target, options=(), channel_credentials=None, call_c
cluster_exp/runtime/rpc_stubs/worker_to_master_pb2_grpc.py:85
MethodExecute
Missing associated documentation comment in .proto file.
cluster_exp/runtime/rpc_stubs/master_to_worker_pb2_grpc.py:42
← previousnext →301–400 of 770, ranked by callers