#!/usr/bin/env python3 import os import json import subprocess from shared import remote_exec, register_job, mongo, upload_code_package code_package = upload_code_package() gpus_per_node = 4 num_workers = 16 experiment = os.path.splitext(os.path.basename(__file__))[0] description = "Repetitions" base_config = { "seed": 1, "task": "Cifar", "model_name": "VGG-11", "algorithm": "relaysum-model", "overlap_communication": False, "base_optimizer": "SGD", "num_epochs": 200, "num_lr_warmup_epochs": 5, "lr_schedule_milestones": [(150, 0.1), (180, 0.1)], "batch_size": 32, "weight_decay": 1e-4, "data_split_method": "dirichlet", "non_iid_alpha": None, "distributed_world_size": num_workers, "gpus_per_node": gpus_per_node, } best_lrs = { (1, "double-binary-trees", 0.0): 0.2 * 2 * 3, (.1, "double-binary-trees", 0.0): 0.2 * 3, (.01, "double-binary-trees", 0.0): 0.2 / 2 * 3, (1, "double-binary-trees", 0.9): 0.2 * 3, (.1, "double-binary-trees", 0.9): 0.1 * 3, (.01, "double-binary-trees", 0.9): 0.1 * 3 / 2, (1, "chain", 0.0): 0.4 * 5, (.1, "chain", 0.0): 0.4 * 5, (.01, "chain", 0.0): 0.2 * 5, (1, "chain", 0.9): 0.1 * 5, (.1, "chain", 0.9): 0.1 * 5, (.01, "chain", 0.9): 0.025 * 5, } for seed in [1, 2, 3]: for alpha in [0.01]: for topology in ["double-binary-trees"]: lr = best_lrs[alpha, topology, 0.9] config = {**base_config, "learning_rate": lr, "momentum": 0.9, "topology": topology, "non_iid_alpha": alpha, "seed": seed} job_name = "alpha{non_iid_alpha}-{algorithm}-{topology}-mom{momentum}-lr{learning_rate}".format(**config) if mongo.job.count_documents({"job": job_name, "experiment": experiment, **{f"config.{key}": value for key, value in config.items()}}) > 0: # We have this one already continue job_id = register_job( user="anonymized", project="average-routing", experiment=experiment, job=job_name, priority=10, n_workers=num_workers, config_overrides=config, runtime_environment={"clone": {"code_package": code_package}, "script": "train.py"}, annotations={"description": description}, ) remote_exec(f'sbatch --nodes 1 --ntasks {num_workers} --gres gpu:{gpus_per_node} --cpus-per-task 2 --job-name="{job_name}" --wrap="srun jobrun {job_id} --mpi"') # for seed in [2, 3]: # for alpha in [1, 0.1, 0.01]: # for topology in ["chain"]: # # # gossip without momentum # lr = best_lrs[alpha, topology, 0.0] # config = {**base_config, "learning_rate": lr, "momentum": 0.0, "topology": topology, "non_iid_alpha": alpha, "seed": seed} # job_name = "alpha{non_iid_alpha}-{algorithm}-{topology}-mom{momentum}-lr{learning_rate}".format(**config) # if mongo.job.count_documents({"job": job_name, "experiment": experiment, **{f"config.{key}": value for key, value in config.items()}}) > 0: # # We have this one already # continue # job_id = register_job( # user="anonymized", # project="average-routing", # experiment=experiment, # job=job_name, # priority=10, # n_workers=num_workers, # config_overrides=config, # runtime_environment={"clone": {"code_package": code_package}, "script": "train.py"}, # annotations={"description": description}, # ) # remote_exec(f'sbatch --nodes 1 --ntasks {num_workers} --gres gpu:{gpus_per_node} --cpus-per-task 2 --job-name="{job_name}" --wrap="srun jobrun {job_id} --mpi"')