github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-09-07 03:52:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

[test] refactored with the new rerun decorator (#763)

Browse Source 
* [test] refactored with the new rerun decorator

* polish test case
This commit is contained in:
Frank Lee
2022-04-15 00:33:04 +08:00
committed by GitHub
parent deaf99f4c9
commit 5a1a095b92
34 changed files with 80 additions and 75 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/test_moe/test_moe_zero_optim.py
View File

@@ -6,7 +6,7 @@ import torch
import torch.multiprocessing as mp
from colossalai.amp import convert_to_apex_amp
from colossalai.nn.optimizer import CPUAdam
from colossalai.testing import parameterize, rerun_on_exception
from colossalai.testing import parameterize, rerun_if_address_is_in_use
from colossalai.utils import free_port
from colossalai.zero.init_ctx import ZeroInitContext
from colossalai.zero.shard_utils import (BucketTensorShardStrategy, TensorShardStrategy)
@@ -120,7 +120,7 @@ def _run_dist(rank, world_size, port):
# use_cpuadam = True can be used with cpu_offload = False
@pytest.mark.dist
@pytest.mark.parametrize("world_size", [2])
@rerun_on_exception(exception_type=mp.ProcessRaisedException, pattern=".*Address already in use.*")
@rerun_if_address_is_in_use()
def test_moe_zero_optim(world_size):
run_func = partial(_run_dist, world_size=world_size, port=free_port())
mp.spawn(run_func, nprocs=world_size)