fix(train): correct --output to --output-dir arg, add Prometheus annotations
The train_baseline_rl binary expects --output-dir, not --output. Also adds prometheus.io scrape annotations to job pod templates so training metrics appear in Grafana. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
@@ -157,7 +157,7 @@ generate_job_manifest() {
|
||||
local cli_args=("--model" "$model" "--symbol" "$SYMBOL" "--data-dir" "/data/futures-baseline")
|
||||
cli_args+=("--mbp10-data-dir" "/data/futures-baseline-mbp10")
|
||||
cli_args+=("--trades-data-dir" "/data/futures-baseline-trades")
|
||||
cli_args+=("--output" "/output")
|
||||
cli_args+=("--output-dir" "/output")
|
||||
if [[ "$PRESET" == "hyperopt" ]]; then
|
||||
cli_args+=("--trials" "$TRIALS")
|
||||
fi
|
||||
@@ -186,6 +186,10 @@ spec:
|
||||
ttlSecondsAfterFinished: 600
|
||||
template:
|
||||
metadata:
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "9094"
|
||||
prometheus.io/path: "/metrics"
|
||||
labels:
|
||||
app.kubernetes.io/part-of: foxhunt
|
||||
app.kubernetes.io/component: training-workflow
|
||||
|
||||
Reference in New Issue
Block a user