Add pipeline tuner component and dependencies. (#671)

* add pipeline tuner component and dependencies.

* clean code.

* do not need force rerun.

* replace the resources.

* update metrics retrieving.

* Update test/pipeline_tuning_example/requirements.txt

* Update test/pipeline_tuning_example/train/env.yaml

* Update test/pipeline_tuning_example/tuner/env.yaml

* Update test/pipeline_tuning_example/tuner/tuner_func.py

* Update test/pipeline_tuning_example/data_prep/env.yaml

* fix issues found by lint with flake8.

* add documentation

* add data.

* do not need AML resource for local run.

* AML -> AzureML

* clean code.

* Update website/docs/Examples/Tune-AzureML pipeline.md

* rename and add pip install.

* update figure name.

* align docs with code.

* remove extra line.

test/pipeline_tuning_example/data_prep/data_prep.py

@@ -0,0 +1,38 @@
+import os
+import argparse
+import pandas as pd
+from sklearn.model_selection import train_test_split
+import logging
+
+logger = logging.getLogger(__name__)
+
+
+def main():
+    """Main function of the script."""
+
+    # input and output arguments
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--data", type=str, help="path to input data")
+    parser.add_argument("--test_train_ratio", type=float, required=False, default=0.25)
+    parser.add_argument("--train_data", type=str, help="path to train data")
+    parser.add_argument("--test_data", type=str, help="path to test data")
+    args = parser.parse_args()
+
+    logger.info(" ".join(f"{k}={v}" for k, v in vars(args).items()))
+
+    data_path = os.path.join(args.data, 'data.csv')
+    df = pd.read_csv(data_path)
+
+    train_df, test_df = train_test_split(
+        df,
+        test_size=args.test_train_ratio,
+    )
+
+    # output paths are mounted as folder, therefore, we are adding a filename to the path
+    train_df.to_csv(os.path.join(args.train_data, "data.csv"), index=False)
+
+    test_df.to_csv(os.path.join(args.test_data, "data.csv"), index=False)
+
+
+if __name__ == "__main__":
+    main()

test/pipeline_tuning_example/data_prep/data_prep.yaml

@@ -0,0 +1,26 @@
+$schema: https://componentsdk.azureedge.net/jsonschema/CommandComponent.json
+name: data_prep
+version: 0.0.1
+display_name: Data preparation for training
+type: CommandComponent
+inputs:
+  data:
+    type: path
+  test_train_ratio:
+    type: float
+outputs:
+  train_data:
+    type: path
+  test_data:
+    type: path
+environment:
+  conda:
+    conda_dependencies_file: env.yaml
+  os: Linux
+
+command: >-
+  python data_prep.py 
+  --data {inputs.data}
+  --test_train_ratio {inputs.test_train_ratio}
+  --train_data {outputs.train_data}
+  --test_data {outputs.test_data}

test/pipeline_tuning_example/data_prep/env.yaml

@@ -0,0 +1,15 @@
+name: data-prep-env
+channels:
+  - conda-forge
+dependencies:
+  - python=3.8
+  - numpy=1.21.2
+  - pip=21.2.4
+  - scikit-learn=0.24.2
+  - scipy=1.7.1
+  - pandas>=1.1,<1.2
+  - pip:
+    # - inference-schema[numpy-support]==1.3.0
+    # - xlrd==2.0.1
+    - mlflow==1.26.1
+    - azureml-mlflow==1.42.0