Skip to content
Merged
Changes from all commits
Commits
Show all changes
21 commits
Select commit Hold shift + click to select a range
96eecb4
Add a deterministic_random reward type
fzyzcjy Jun 22, 2026
2c1d4a7
Add an inplace_modify_args context manager
fzyzcjy Jun 22, 2026
7b00c75
Add fault-tolerance support tweaks to shared utilities
fzyzcjy Jun 22, 2026
635aa50
Preserve the process-group backend across reload
fzyzcjy Jun 22, 2026
7a33a63
Add fault-tolerance foundation utilities
fzyzcjy Jun 22, 2026
8dc8ddd
Add structured logfmt logging helper
fzyzcjy Jun 22, 2026
96d220f
Add a Clock abstraction with a fake clock for tests
fzyzcjy Jun 22, 2026
a005a2f
Add a fault injector test utility
fzyzcjy Jun 22, 2026
8365971
Add control-server data models
fzyzcjy Jun 22, 2026
e308821
Add a cell health checker and heartbeat utilities
fzyzcjy Jun 22, 2026
4a52fcf
Add the fault-tolerance dependency, CI label, and logger-config setup
fzyzcjy Jun 22, 2026
2ed9fcf
Always reconnect rollout engines on weight-update setup
fzyzcjy Jun 22, 2026
a838a0b
Add a fault-injection RPC to train actors
fzyzcjy Jun 22, 2026
c7798b5
Delay splitting train data by DP until actor-side processing
fzyzcjy Jul 8, 2026
d0b41fc
Add a deterministic NCCL backend for order-stable collectives
fzyzcjy Jun 22, 2026
a48d03f
Add a per-process identity helper
fzyzcjy Jun 22, 2026
3e85a86
Add structured event models keyed by per-process identity
fzyzcjy Jun 22, 2026
dc11be3
Add structured event logging keyed by per-process identity
fzyzcjy Jun 22, 2026
691e4a7
Add event-log snapshot and restore checkpointing
fzyzcjy Jun 22, 2026
5a91150
Log training metrics as MetricEvents through the event logger
fzyzcjy Jun 22, 2026
d501c71
Merge remote-tracking branch 'origin/main' into tom/pr_chain/trainer_…
fzyzcjy Jul 10, 2026
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 9 additions & 0 deletions miles/utils/tracking_utils/__init__.py
Original file line number Diff line number Diff line change
@@ -1,5 +1,10 @@
import logging

import torch

from miles.utils.event_logger.logger import get_event_logger, is_event_logger_initialized
from miles.utils.event_logger.models import MetricEvent

from .base import TrackingManager

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nit: let's also fix this relative import.


logger = logging.getLogger(__name__)
Expand All @@ -14,6 +19,10 @@ def log(args, metrics, step_key: str):
step = metrics.get(step_key)
_manager.log(metrics, step=step, step_key=step_key)

if is_event_logger_initialized():
serializable_metrics = {k: (v.item() if isinstance(v, torch.Tensor) else v) for k, v in metrics.items()}
get_event_logger().log(MetricEvent, {"metrics": serializable_metrics}, print_log=False)


def finish_tracking():
_manager.finish()
Loading