gitea_admin/sientia-dataops-model-manager
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

SIENTIAPDE-1241: refactor train_model workflow due to I/O errors.

Browse Source
This commit is contained in:
Bruno Domingues
2025-10-22 15:37:56 -03:00
parent f2a1c88ff3
commit 5789a13023
31 changed files with 37878 additions and 6097 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
.env.example
View File

@@ -6,8 +6,7 @@ POSTGRES_DBNAME="sientia"
POSTGRES_MIN_CONNECTIONS="10"
POSTGRES_MAX_CONNECTIONS="30"
MLFLOW_HOST="http://sientia-tracker-mlflow-tracking.sientia-tracker.svc.cluster.local"
MLFLOW_PORT="80"
MLFLOW_URL="http://sientia-tracker-mlflow-tracking.sientia-tracker.svc.cluster.local:80"
MLFLOW_USERNAME="aignosi"
MLFLOW_PASSWORD="mlflow_password"
@@ -38,11 +37,8 @@ MINIO_READ_TIMEOUT="60"
# Workflow Activity Timeouts (in seconds)
# These timeouts are designed to handle large files (up to 200MB)
TIMEOUT_VALIDATE_PARAMS="30" # Parameter validation (fast operation)
TIMEOUT_DOWNLOAD_FILE="600" # File download from MinIO (10 min for 200MB @ 1MB/s with 3x buffer)
TIMEOUT_TRAIN_MODEL="1800" # Model training (30 min for large datasets)
TIMEOUT_SAVE_MODEL="300" # Save model to MLFlow (5 min for artifacts upload)
TIMEOUT_CLEANUP_DIRECTORY="60" # Cleanup temporary directory (1 min)
TIMEOUT_DELETE_FILE="60" # Delete file from MinIO (1 min)
TIMEOUT_TRAIN_MODEL="2700" # Model training (30 min for large datasets)
TIMEOUT_DELETE_FILE="120" # Delete file from MinIO (1 min)
TIMEOUT_UPDATE_DATABASE="30" # Database update operations (30 sec)
EXTRA_PIP_REQUIREMENTS="git+https://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@github.com/Aignosi/sientia-mlops-library.git"