Implementation:Norrrrrrr lyn WAInjectBench LogisticRegression Fit

Knowledge Sources	WAInjectBench Scikit-learn
Domains	Machine_Learning, Classification
Last Updated	2026-02-14 16:00 GMT

Overview

Concrete tool for training binary logistic regression classifiers on embedding features, provided by scikit-learn as used in the WAInjectBench embedding trainers.

Description

Both text and image embedding trainers use sklearn.linear_model.LogisticRegression to fit a binary classifier. The text variant uses max_iter=1000; the image variant uses max_iter=2000, class_weight="balanced", and n_jobs=-1. After fitting, both print a classification_report on the training data.

Usage

Called after embedding extraction to fit a classifier on the feature matrix and label vector.

Code Reference

Source Location

Repository: WAInjectBench
File: train/embedding-t.py (L30-31), train/embedding-i.py (L48-53)

Signature

# Text variant (train/embedding-t.py:L30-31)
clf = LogisticRegression(max_iter=1000)
clf.fit(embeddings, labels)

# Image variant (train/embedding-i.py:L48-53)
clf = LogisticRegression(
    max_iter=2000,
    class_weight="balanced",
    n_jobs=-1
)
clf.fit(embeddings, labels)

# Both variants follow with:
preds = clf.predict(embeddings)
print(classification_report(labels, preds))

Import

from sklearn.linear_model import LogisticRegression
from sklearn.metrics import classification_report

I/O Contract

Inputs

Name	Type	Required	Description
embeddings	np.ndarray	Yes	Feature matrix of shape (N, embedding_dim)
labels	List[int]	Yes	Binary labels (0=benign, 1=malicious)

Outputs

Name	Type	Description
clf	LogisticRegression	Fitted classifier object ready for predict() calls
classification_report	str (printed)	Precision/recall/F1 on training data

Usage Examples

Training a Text Classifier

from sklearn.linear_model import LogisticRegression
from sklearn.metrics import classification_report
import numpy as np

# Assume embeddings: np.ndarray (N, 384), labels: List[int]
clf = LogisticRegression(max_iter=1000)
clf.fit(embeddings, labels)

preds = clf.predict(embeddings)
print(classification_report(labels, preds))

Training an Image Classifier with Balanced Weights

clf = LogisticRegression(max_iter=2000, class_weight="balanced", n_jobs=-1)
clf.fit(embeddings, labels)

preds = clf.predict(embeddings)
print(classification_report(labels, preds))

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment