MLflow 入門
Weston Bassler
Senior MLOps Engineer

模型版本
模型階段

載入模型
# MLflow flavor
mlflow.FLAVOR.load_model()
服務模型
# MLflow serve 指令列
mlflow models serve
慣例
models:/
模型版本
models:/model_name/version
模型階段
models:/model_name/stage
# 匯入 flavor import mlflow.FLAVOR# 載入版本 mlflow.FLAVOR.load_model("models:/model_name/version")# 載入階段 mlflow.FLAVOR.load_model("models:/model_name/stage")
# 匯入 flavor import mlflow.sklearn# 載入 Staging 的 Unicorn 模型 model = mlflow.sklearn.load_model("models:/Unicorn/Staging")# 列印模型 model
LogisticRegression()
# 推論
model.predict(data)
# 服務 Production 階段的 Unicorn 模型
mlflow models serve -m "models:/Unicorn/Production"
2023/03/26 15:07:00 INFO mlflow.models.flavor_backend_registry:
Selected backend for flavor 'python_function'
2023/03/26 15:07:00 INFO mlflow.pyfunc.backend: === Running command 'exec gunicorn
--timeout=60 -b 127.0.0.1:5000 -w 1 ${GUNICORN_CMD_ARGS} --
mlflow.pyfunc.scoring_server.wsgi:app'
[2023-03-26 15:07:00 -0400] [86409] [INFO] Starting gunicorn 20.1.0
[2023-03-26 15:07:00 -0400] [86409] [INFO] Listening at: http://127.0.0.1:5000
[2023-03-26 15:07:00 -0400] [86409] [INFO] Using worker: sync
[2023-03-26 15:07:00 -0400] [86410] [INFO] Booting worker with pid: 86410
CSV 格式
pandas_df.to_csv()
JSON 格式
{
"dataframe_split": {
"columns": ["R&D Spend", "Administration", "Marketing Spend", "State"],
"data": [["165349.20", 136897.80, 471784.10, 1]]
}
}
# 將負載送到 invocations 端點
curl http://127.0.0.1:5000/invocations -H 'Content-Type: application/json' -d
{
"dataframe_split": {
"columns": ["R&D Spend", "Administration", "Marketing Spend", "State"],
"data": [["165349.20", 136897.80, 471784.10, 1]]
}
}
[[104055.1842384]]
MLflow 入門