Ksama Arora
Sample code to create a Compenent
Step 1: Script named prep.py to prepare data
# import libraries
import argparse
import pandas as pd
import numpy as np
from pathlib import Path
from sklearn.preprocessing import MinMaxScaler
# setup arg parser
parser = argparse.ArgumentParser()
# add arguments
parser.add_argument("--input_data", dest='input_data',
type=str)
parser.add_argument("--output_data", dest='output_data',
type=str)
# parse args
args = parser.parse_args()
# read the data
df = pd.read_csv(args.input_data)
# remove missing values
df = df.dropna()
# normalize the data
scaler = MinMaxScaler()
num_cols = ['feature1','feature2','feature3','feature4']
df[num_cols] = scaler.fit_transform(df[num_cols])
# save the data as a csv
output_df = df.to_csv(
(Path(args.output_data) / "prepped-data.csv"),
index = False
)
Step 2: Create YAML file
To create component for prep.py script, you will need YAML file prep.yml (stored in src folder)
$schema: https://azuremlschemas.azureedge.net/latest/commandComponent.schema.json
name: prep_data
display_name: Prepare training data
version: 1
type: command
inputs:
input_data:
type: uri_file
outputs:
output_data:
type: uri_file
code: ./src
environment: azureml:AzureML-sklearn-0.24-ubuntu18.04-py37-cpu@latest
command: >-
python prep.py
--input_data $
--output_data $
Step 3: Load Component
from azure.ai.ml import load_component
parent_dir = ""
loaded_component_prep = load_component(source=parent_dir + "./prep.yml")
Step 4: Register Component
When loaded component, register the component.
prep = ml_client.components.create_or_update(prepare_data_component)