Advent of 2022, Day 16 – MLflow in action with xgboost
This article is originally published at https://tomaztsql.wordpress.com
In the series of Azure Machine Learning posts:
- Dec 01: What is Azure Machine Learning?
- Dec 02: Creating Azure Machine Learning Workspace
- Dec 03: Understanding Azure Machine Learning Studio
- Dec 04: Getting data to Azure Machine Learning workspace
- Dec 05: Creating compute and cluster instances in Azure Machine Learning
- Dec 06: Environments in Azure Machine Learning
- Dec 07: Introduction to Azure CLI and Python SDK
- Dec 08: Python SDK namespaces for workspace, experiments and models
- Dec 09: Python SDK namespaces for environment, and pipelines
- Dec 10: Connecting to client using Python SDK namespaces
- Dec 11: Creating Pipelines with Python SDK
- Dec 12: Creating jobs
- Dec 13: Automated ML
- Dec 14: Registering the models
- Dec 15: Getting to know MLflow
Yesterday we have looked into how to start the MLflow configurations and today, let’s put this to the test.
We will create a new notebook and use Heart dataset (link to dataset) to toy around. We will also import xgboost classifier to asses the accuracy of the presence of heart disease in the patient. We will be using a categorical (integer) variable with values from 0 (no presence) to 4 (strong presence) and attempt to classify based on 15+ attributes (out of more than 70 attributes).
The ipynb notebook will be available on the Github.
#importing mlflow functions
import mlflow
mlflow.set_experiment(experiment_name="heart-condition-classifier")
#getting the data
import pandas as pd
file_url = "http://storage.googleapis.com/download.tensorflow.org/data/heart.csv"
df = pd.read_csv(file_url)
#some data engineering
df["thal"] = df["thal"].astype("category").cat.codes
#split train and test
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(
df.drop("target", axis=1), df["target"], test_size=0.4
)
#logging the steps
mlflow.xgboost.autolog()
#training the model
from xgboost import XGBClassifier
model = XGBClassifier(use_label_encoder=False, eval_metric="logloss")
# start the mlflow run
run = mlflow.start_run()
#start fitting the model
model.fit(X_train, y_train, eval_set=[(X_test, y_test)], verbose=False)
#logging some extra metrics
y_pred = model.predict(X_test)
from sklearn.metrics import accuracy_score, recall_score, fbeta_score, confusion_matrix
accuracy = accuracy_score(y_test, y_pred)
recall = recall_score(y_test, y_pred)
cm = confusion_matrix(y_test, y_pred)
#closing mlflow
mlflow.end_run()
run = mlflow.get_run(run.info.run_id)
client = mlflow.tracking.MlflowClient()
We can also do the logging with preprocessing.
# Reload the dataset
df = pd.read_csv(file_url)
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(
df.drop("target", axis=1), df["target"], test_size=0.3
)
#using ordinal encoder
import numpy as np
from sklearn.preprocessing import OrdinalEncoder
# creating transformation and using Logloss on xbgoost classifies
from sklearn.compose import ColumnTransformer
from xgboost import XGBClassifier
encoder = ColumnTransformer(
[
(
"cat_encoding",
OrdinalEncoder(
categories="auto",
handle_unknown="use_encoded_value",
unknown_value=np.nan,
),
["thal"],
)
],
remainder="passthrough",
verbose_feature_names_out=False,
)
model = XGBClassifier(use_label_encoder=False, eval_metric="logloss")
With mulitple runs, you can also check the performance of models with desired metrics. This is an example of logloss validation and comparison between two runs.
Compete set of code, documents, notebooks, and all of the materials will be available at the Github repository: https://github.com/tomaztk/Azure-Machine-Learning
Happy Advent of 2022!
Thanks for visiting r-craft.org
This article is originally published at https://tomaztsql.wordpress.com
Please visit source website for post related comments.
