工作流

MLflow 入门

Weston Bassler

Senior MLOps Engieer

MLflow Projects

工作流

1 unsplash.com
MLflow 入门

MLproject

name: project_name
python_env: python_env.yaml
entry_points:

  step_1:
    command: "python train_model.py"

  step_2:
    command: "python evaluate_model.py {run_id}"
    parameters:
        run_id:
            type: str
            default: None
MLflow 入门

工作流

import mlflow

# Step 1
step_1 = mlflow.projects.run(
    uri='./',
    entry_point='step_1'
)

# Step 2
step_2 = mlflow.projects.run(
    uri='./',
    entry_point='step_2'
)
MLflow 入门

运行 Projects

import mlflow

# Step 1
step_1 = mlflow.projects.run(
    uri='./',
    entry_point='step_1'
)


print(step_1)

<mlflow.projects.submitted_run.LocalSubmittedRun object at 0x125eac8b0>
MLflow 入门

运行 Projects

step_1.cancel() - 终止当前运行

step_1.get_status() - 获取运行状态

step_1.run_id - 该运行的 run_id

step_1.wait() - 等待运行完成

MLflow 入门

运行 Projects

import mlflow

# Step 1
step_1 = mlflow.projects.run(
    uri='./',
    entry_point='step_1'
)


# 为 step_1 设置 run_id 变量
step_1_run_id = step_1.run_id

# Step 2
step_2 = mlflow.projects.run(
    uri='./',
    entry_point='step_2',

    parameters={
        'run_id': step_1_run_id
    }

)
MLflow 入门

ML 生命周期

模型工程与模型评估

MLflow 入门

¡Vamos a practicar!

MLflow 入门

Preparing Video For Download...