Skip to content

Commit

Permalink
.
Browse files Browse the repository at this point in the history
  • Loading branch information
garrett4wade committed Sep 2, 2024
1 parent 6b9904d commit 81b45eb
Showing 1 changed file with 2 additions and 1 deletion.
3 changes: 2 additions & 1 deletion realhf/base/constants.py
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
import pathlib
from collections import defaultdict
from typing import *
import datetime

import numpy as np

Expand Down Expand Up @@ -56,7 +57,7 @@ def get_tensor(self, tensor_shape, dtype, name, force_zero: bool = False):
# 30 minutes. Transferring super-large batches via NCCL bcast
# for the first time may consumer over 600 secs, which is the
# pytorch's default. Increase this value to 30 minutes.
NCCL_DEFAULT_TIMEOUT = 1800
NCCL_DEFAULT_TIMEOUT = datetime.timedelta(seconds=1800)

# constants in experiment instance scope
MODEL_SAVE_ROOT = f"{cluster_spec.fileroot}/checkpoints/{getpass.getuser()}"
Expand Down

0 comments on commit 81b45eb

Please sign in to comment.