Regression on the hotel reviews [scikit-learn]¶
Giskard is an open-source framework for testing all ML models, from LLMs to tabular models. Don’t hesitate to give the project a star on GitHub ⭐️ if you find it useful!
In this notebook, you’ll learn how to create comprehensive test suites for your model in a few lines of code, thanks to Giskard’s open-source Python library.
Use-case:
Regression task of predicting review ‘score’, based on the review text.
Outline:
Detect vulnerabilities automatically with Giskard’s scan
Automatically generate & curate a comprehensive test suite to test your model beyond accuracy-related metrics
Upload your model to the Giskard Hub to:
Debug failing tests & diagnose issues
Compare models & decide which one to promote
Share your results & collect feedback from non-technical team members
Install dependencies¶
Make sure to install the giskard
[ ]:
%pip install giskard --upgrade
Import libraries¶
[1]:
from pathlib import Path
from urllib.request import urlretrieve
import pandas as pd
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics import mean_absolute_error
from sklearn.model_selection import train_test_split
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import FunctionTransformer
from typing import Iterable
from giskard import Model, Dataset, scan, testing, GiskardClient, Suite
Define constants¶
[2]:
# Constants.
FEATURE_COLUMN_NAME = "Full_Review"
TARGET_COLUMN_NAME = "Reviewer_Score"
# Paths.
DATA_URL = "ftp://sys.giskard.ai/pub/unit_test_resources/hotel_text_regression_dataset/Hotel_Reviews.csv"
DATA_PATH = Path.home() / ".giskard" / "hotel_text_regression_dataset" / "Hotel_Reviews.csv"
Dataset preparation¶
Load data¶
[3]:
def fetch_from_ftp(url: str, file: Path) -> None:
"""Helper to fetch data from the FTP server."""
if not file.parent.exists():
file.parent.mkdir(parents=True, exist_ok=True)
if not file.exists():
print(f"Downloading data from {url}")
urlretrieve(url, file)
print(f"Data was loaded!")
def load_data(**kwargs) -> pd.DataFrame:
fetch_from_ftp(DATA_URL, DATA_PATH)
df = pd.read_csv(DATA_PATH, **kwargs)
# Create target column.
df[FEATURE_COLUMN_NAME] = df.apply(lambda x: x['Positive_Review'] + ' ' + x['Negative_Review'], axis=1)
return df
[ ]:
reviews_df = load_data(nrows=1000)
Train-test split¶
[5]:
train_X, test_X, train_Y, test_Y = train_test_split(reviews_df[[FEATURE_COLUMN_NAME]], reviews_df[TARGET_COLUMN_NAME],
random_state=42)
Wrap dataset with Giskard¶
To prepare for the vulnerability scan, make sure to wrap your dataset using Giskard’s Dataset class. More details here.
[6]:
raw_data = pd.concat([test_X, test_Y], axis=1)
giskard_dataset = Dataset(
df=raw_data,
# A pandas.DataFrame that contains the raw data (before all the pre-processing steps) and the actual ground truth variable (target).
target=TARGET_COLUMN_NAME, # Ground truth variable.
name="hotel_text_regression_dataset", # Optional.
)
Model building¶
Define preprocessing steps¶
[7]:
def adapt_vectorizer_input(df: pd.DataFrame) -> Iterable:
"""Adapt input for the vectorizers.
The problem is that vectorizers accept iterable, not DataFrame, but Series.
Thus, we need to ravel dataframe with text have input single dimension.
"""
df = df.iloc[:, 0]
return df
Build estimator¶
[ ]:
# Define pipeline.
pipeline = Pipeline(steps=[
("vectorizer_adapter", FunctionTransformer(adapt_vectorizer_input)),
("vectorizer", TfidfVectorizer(max_features=10000)),
("regressor", GradientBoostingRegressor(n_estimators=10))
])
# Fit pipeline.
pipeline.fit(train_X, train_Y)
# Perform inference on train and test data.
pred_train = pipeline.predict(train_X)
pred_test = pipeline.predict(test_X)
train_metric = mean_absolute_error(train_Y, pred_train)
test_metric = mean_absolute_error(test_Y, pred_test)
print(f"Train MAE: {train_metric: .2f}\n"
f"Test MAE: {test_metric: .2f}")
Wrap model with Giskard¶
To prepare for the vulnerability scan, make sure to wrap your model using Giskard’s Model class. You can choose to either wrap the prediction function (preferred option) or the model object. More details here.
[ ]:
# Wrap the prediction method to allow saving the whole pipeline to the Hub
def prediction_function(df):
return pipeline.predict(df)
giskard_model = Model(
model=prediction_function,
# A prediction function that encapsulates all the data pre-processing steps and that could be executed with the dataset used by the scan.
model_type="regression", # Either regression, classification or text_generation.
name="hotel_text_regression", # Optional.
feature_names=[FEATURE_COLUMN_NAME] # Default: all columns of your dataset.
)
# Validate wrapped model.
pred_test_wrapped = giskard_model.predict(giskard_dataset).raw_prediction
wrapped_test_metric = mean_absolute_error(test_Y, pred_test_wrapped)
print(f"Wrapped Test MAE: {wrapped_test_metric: .2f}")
Detect vulnerabilities in your model¶
Scan your model for vulnerabilities with Giskard¶
Giskard’s scan allows you to detect vulnerabilities in your model automatically. These include performance biases, unrobustness, data leakage, stochasticity, underconfidence, ethical issues, and more. For detailed information about the scan feature, please refer to our scan documentation.
[ ]:
results = scan(giskard_model, giskard_dataset)
[11]:
display(results)
Generate comprehensive test suites automatically for your model¶
Generate test suites from the scan¶
The objects produced by the scan can be used as fixtures to generate a test suite that integrate all detected vulnerabilities. Test suites allow you to evaluate and validate your model’s performance, ensuring that it behaves as expected on a set of predefined test cases, and to identify any regressions or issues that might arise during development or updates.
[12]:
test_suite = results.generate_test_suite("My first test suite")
test_suite.run()
Executed 'Invariance to “Add typos”' with arguments {'model': <giskard.models.function.PredictionFunctionModel object at 0x12af83400>, 'dataset': <giskard.datasets.base.Dataset object at 0x12acc2b00>, 'transformation_function': <giskard.scanner.robustness.text_transformations.TextTypoTransformation object at 0x12acc1480>, 'threshold': 0.95, 'output_sensitivity': 0.05}:
Test failed
Metric: 0.87
- [TestMessageLevel.INFO] 239 rows were perturbed
Executed 'MSE on data slice “`Full_Review` contains "building"”' with arguments {'model': <giskard.models.function.PredictionFunctionModel object at 0x12af83400>, 'dataset': <giskard.datasets.base.Dataset object at 0x12acc2b00>, 'slicing_function': <giskard.slicing.slice.QueryBasedSliceFunction object at 0x12eba9cc0>, 'threshold': 2.3372302706019896}:
Test failed
Metric: 3.6
Executed 'MSE on data slice “`Full_Review` contains "stay"”' with arguments {'model': <giskard.models.function.PredictionFunctionModel object at 0x12af83400>, 'dataset': <giskard.datasets.base.Dataset object at 0x12acc2b00>, 'slicing_function': <giskard.slicing.slice.QueryBasedSliceFunction object at 0x12ebed2a0>, 'threshold': 2.3372302706019896}:
Test failed
Metric: 3.4
Executed 'MSE on data slice “`Full_Review` contains "bed"”' with arguments {'model': <giskard.models.function.PredictionFunctionModel object at 0x12af83400>, 'dataset': <giskard.datasets.base.Dataset object at 0x12acc2b00>, 'slicing_function': <giskard.slicing.slice.QueryBasedSliceFunction object at 0x12ec508b0>, 'threshold': 2.3372302706019896}:
Test failed
Metric: 2.96
Executed 'MSE on data slice “`Full_Review` contains "comfy"”' with arguments {'model': <giskard.models.function.PredictionFunctionModel object at 0x12af83400>, 'dataset': <giskard.datasets.base.Dataset object at 0x12acc2b00>, 'slicing_function': <giskard.slicing.slice.QueryBasedSliceFunction object at 0x12ec52ef0>, 'threshold': 2.3372302706019896}:
Test failed
Metric: 2.72
Executed 'MSE on data slice “`Full_Review` contains "area"”' with arguments {'model': <giskard.models.function.PredictionFunctionModel object at 0x12af83400>, 'dataset': <giskard.datasets.base.Dataset object at 0x12acc2b00>, 'slicing_function': <giskard.slicing.slice.QueryBasedSliceFunction object at 0x12ecd0c40>, 'threshold': 2.3372302706019896}:
Test failed
Metric: 2.63
Executed 'MSE on data slice “`Full_Review` contains "food"”' with arguments {'model': <giskard.models.function.PredictionFunctionModel object at 0x12af83400>, 'dataset': <giskard.datasets.base.Dataset object at 0x12acc2b00>, 'slicing_function': <giskard.slicing.slice.QueryBasedSliceFunction object at 0x12eca9fc0>, 'threshold': 2.3372302706019896}:
Test failed
Metric: 2.57
Executed 'MSE on data slice “`Full_Review` contains "hotel"”' with arguments {'model': <giskard.models.function.PredictionFunctionModel object at 0x12af83400>, 'dataset': <giskard.datasets.base.Dataset object at 0x12acc2b00>, 'slicing_function': <giskard.slicing.slice.QueryBasedSliceFunction object at 0x12eb97f40>, 'threshold': 2.3372302706019896}:
Test failed
Metric: 2.52
Executed 'MSE on data slice “`Full_Review` contains "room"”' with arguments {'model': <giskard.models.function.PredictionFunctionModel object at 0x12af83400>, 'dataset': <giskard.datasets.base.Dataset object at 0x12acc2b00>, 'slicing_function': <giskard.slicing.slice.QueryBasedSliceFunction object at 0x12ec7e6b0>, 'threshold': 2.3372302706019896}:
Test failed
Metric: 2.4
[12]:
Customize your suite by loading objects from the Giskard catalog¶
The Giskard open source catalog will enable to load:
Tests such as metamorphic, performance, prediction & data drift, statistical tests, etc
Slicing functions such as detectors of toxicity, hate, emotion, etc
Transformation functions such as generators of typos, paraphrase, style tune, etc
To create custom tests, refer to this page.
For demo purposes, we will load a simple unit test (test_r2) that checks if the test R2 score is above the given threshold. For more examples of tests and functions, refer to the Giskard catalog.
[ ]:
test_suite.add_test(testing.test_r2(model=giskard_model, dataset=giskard_dataset, threshold=0.7)).run()
Debug and interact with your tests in the Giskard Hub¶
At this point, you’ve created a test suite that is highly specific to your domain & use-case. Failing tests can be a pain to debug, which is why we encourage you to head over to the Giskard Hub.
Play around with a demo of the Giskard Hub on HuggingFace Spaces using this link.
More than just debugging tests, the Giskard Hub allows you to:
Compare models to decide which model to promote
Automatically create additional domain-specific tests through our automated model insights feature
Share your test results with team members and decision makers
The Giskard Hub can be deployed easily on HuggingFace Spaces.
Here’s a sneak peek of automated model insights on a credit scoring classification model.
Upload your test suite to the Giskard Hub¶
The entry point to the Giskard Hub is the upload of your test suite. Uploading the test suite will automatically save the model, dataset, tests, slicing & transformation functions to the Giskard Hub.
[ ]:
# Create a Giskard client after having install the Giskard server (see documentation)
api_key = "<Giskard API key>" #This can be found in the Settings tab of the Giskard hub
#hf_token = "<Your Giskard Space token>" #If the Giskard Hub is installed on HF Space, this can be found on the Settings tab of the Giskard Hub
client = GiskardClient(
url="http://localhost:19000", # Option 1: Use URL of your local Giskard instance.
# url="<URL of your Giskard hub Space>", # Option 2: Use URL of your remote HuggingFace space.
key=api_key,
# hf_token=hf_token # Use this token to access a private HF space.
)
project_key = "my_project"
my_project = client.create_project(project_key, "PROJECT_NAME", "DESCRIPTION")
# Upload to the project you just created
test_suite.upload(client, project_key)
Download a test suite from the Giskard Hub¶
After curating your test suites with additional tests on the Giskard Hub, you can easily download them back into your environment. This allows you to:
Check for regressions after training a new model
Automate the test suite execution in a CI/CD pipeline
Compare several models during the prototyping phase
[ ]:
test_suite_downloaded = Suite.download(client, project_key, suite_id=...)
test_suite_downloaded.run()