fix(training): disable DataLoader workers in subprocess training
DataLoader worker subprocesses crash inside Docker due to multiprocessing fork restrictions. Pass --workers 0 to both ketos train and ketos segtrain so data loading runs in the main process. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
@@ -366,7 +366,8 @@ async def train_model(
|
||||
os.makedirs(checkpoint_dir, exist_ok=True)
|
||||
|
||||
cmd = [
|
||||
"ketos", "train",
|
||||
"ketos", "--workers", "0",
|
||||
"train",
|
||||
"-f", "path",
|
||||
"-o", checkpoint_dir,
|
||||
"-q", "fixed",
|
||||
@@ -456,7 +457,8 @@ async def segtrain_model(
|
||||
os.makedirs(checkpoint_dir, exist_ok=True)
|
||||
|
||||
cmd = [
|
||||
"ketos", "segtrain",
|
||||
"ketos", "--workers", "0",
|
||||
"segtrain",
|
||||
"-o", checkpoint_dir,
|
||||
"-q", "fixed",
|
||||
"-N", "50",
|
||||
|
||||
Reference in New Issue
Block a user