NASP/argus-cluster
84
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

V3.6 增加W&B server，以及配置任务默认都开wandb

Browse Source
This commit is contained in:
yuyr
2026-01-05 17:04:10 +08:00
parent 70f52a907b
commit a6c66995e3
25 changed files with 5852 additions and 2223 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
src/mvp/py/argus/service/app.py
View File

@@ -276,6 +276,14 @@ def create_app(config_path: str) -> FastAPI:
"jobs_purge_after_days": int(v2_cfg.data.retention.jobs_purge_after_days),
},
}
if v2_cfg.tracking.wandb.enabled:
suffix = v2_cfg.tracking.wandb.project_suffix or "_project"
out["wandb"] = {
"enabled": True,
# Training uses base_url; UI can also open the host port directly.
"base_url": v2_cfg.tracking.wandb.base_url,
"project_name": f"{user_id}{suffix}",
}
if _sftpgo_enabled():
out["sftp"] = {
"host": v2_cfg.data.sftpgo.host,
@@ -429,7 +437,7 @@ def create_app(config_path: str) -> FastAPI:
"error_summary": row.get("error_summary"),
}
if latest_attempt:
out["latest_attempt"] = {
latest_out: dict[str, Any] = {
"attempt_no": latest_attempt["attempt_no"],
"ray_submission_id": latest_attempt["ray_submission_id"],
"ray_status": latest_attempt.get("ray_status"),
@@ -438,6 +446,16 @@ def create_app(config_path: str) -> FastAPI:
"failure_kind": latest_attempt.get("failure_kind"),
"message": latest_attempt.get("message"),
}
if v2_cfg.tracking.wandb.enabled:
suffix = v2_cfg.tracking.wandb.project_suffix or "_project"
# For v3.6: shared W&B account, user-level projects.
# Use attempt's ray_submission_id as run name for stable mapping.
latest_out["wandb"] = {
"base_url": v2_cfg.tracking.wandb.base_url,
"project_name": f"{str(row.get('user_id') or subject.get('user_id'))}{suffix}",
"run_name": str(latest_attempt.get("ray_submission_id") or ""),
}
out["latest_attempt"] = latest_out
return out
@app.get("/api/v2/tasks/{task_id}/spec")