Spaces:
Sleeping
Sleeping
File size: 4,479 Bytes
2895c00 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 |
# Copyright (c) Facebook, Inc. and its affiliates.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
A script to run multinode training with submitit.
Almost copy-paste from https://github.com/facebookresearch/deit/blob/main/run_with_submitit.py
"""
import argparse
import os
import uuid
from pathlib import Path
import main_dino
import submitit
def parse_args():
parser = argparse.ArgumentParser(
"Submitit for DINO", parents=[main_dino.get_args_parser()]
)
parser.add_argument(
"--ngpus", default=8, type=int, help="Number of gpus to request on each node"
)
parser.add_argument(
"--nodes", default=2, type=int, help="Number of nodes to request"
)
parser.add_argument("--timeout", default=2800, type=int, help="Duration of the job")
parser.add_argument(
"--partition", default="learnfair", type=str, help="Partition where to submit"
)
parser.add_argument(
"--use_volta32", action="store_true", help="Big models? Use this"
)
parser.add_argument(
"--comment",
default="",
type=str,
help="Comment to pass to scheduler, e.g. priority message",
)
return parser.parse_args()
def get_shared_folder() -> Path:
user = os.getenv("USER")
if Path("/checkpoint/").is_dir():
p = Path(f"/checkpoint/{user}/experiments")
p.mkdir(exist_ok=True)
return p
raise RuntimeError("No shared folder available")
def get_init_file():
# Init file must not exist, but it's parent dir must exist.
os.makedirs(str(get_shared_folder()), exist_ok=True)
init_file = get_shared_folder() / f"{uuid.uuid4().hex}_init"
if init_file.exists():
os.remove(str(init_file))
return init_file
class Trainer(object):
def __init__(self, args):
self.args = args
def __call__(self):
import main_dino
self._setup_gpu_args()
main_dino.train_dino(self.args)
def checkpoint(self):
import os
import submitit
self.args.dist_url = get_init_file().as_uri()
print("Requeuing ", self.args)
empty_trainer = type(self)(self.args)
return submitit.helpers.DelayedSubmission(empty_trainer)
def _setup_gpu_args(self):
import submitit
from pathlib import Path
job_env = submitit.JobEnvironment()
self.args.output_dir = Path(
str(self.args.output_dir).replace("%j", str(job_env.job_id))
)
self.args.gpu = job_env.local_rank
self.args.rank = job_env.global_rank
self.args.world_size = job_env.num_tasks
print(f"Process group: {job_env.num_tasks} tasks, rank: {job_env.global_rank}")
def main():
args = parse_args()
if args.output_dir == "":
args.output_dir = get_shared_folder() / "%j"
Path(args.output_dir).mkdir(parents=True, exist_ok=True)
executor = submitit.AutoExecutor(folder=args.output_dir, slurm_max_num_timeout=30)
num_gpus_per_node = args.ngpus
nodes = args.nodes
timeout_min = args.timeout
partition = args.partition
kwargs = {}
if args.use_volta32:
kwargs["slurm_constraint"] = "volta32gb"
if args.comment:
kwargs["slurm_comment"] = args.comment
executor.update_parameters(
mem_gb=40 * num_gpus_per_node,
gpus_per_node=num_gpus_per_node,
tasks_per_node=num_gpus_per_node, # one task per GPU
cpus_per_task=10,
nodes=nodes,
timeout_min=timeout_min, # max is 60 * 72
# Below are cluster dependent parameters
slurm_partition=partition,
slurm_signal_delay_s=120,
**kwargs,
)
executor.update_parameters(name="dino")
args.dist_url = get_init_file().as_uri()
trainer = Trainer(args)
job = executor.submit(trainer)
print(f"Submitted job_id: {job.job_id}")
print(f"Logs and checkpoints will be saved at: {args.output_dir}")
if __name__ == "__main__":
main()
|