Full API

Classifier

Use when your target column is a category (yes/no, species, label). Wraps scikit-learn classifiers with automatic preprocessing.

Constructor

python

from fling import Classifier

model = Classifier(data, target, test_size=0.2, random_state=42)

ℹPrints a loading confirmation. Does not preprocess yet - preprocessing happens in .train().

Parameters

Name	Type	Description
`data`	str or DataFrame	Path to a CSV file, or a pandas DataFrame
`target`	str	Name of the column to predict
`test_size`	float	Fraction of data held out for evaluation (default `0.2`)
`random_state`	int	Random seed for reproducibility (default `42`)

Algorithm strings

Pass to .train(algorithm=...). Omit to let fling auto-select.

String	Model
`"logistic"`	LogisticRegression
`"random_forest"`	RandomForestClassifier
`"gradient_boosting"`	GradientBoostingClassifier
`"decision_tree"`	DecisionTreeClassifier
`"knn"`	KNeighborsClassifier
`"svm"`	SVC

Auto-selection logic

Dataset size	Default model	Why
< 1,000 rows	LogisticRegression	Fast, interpretable on small data
1,000 – 50,000 rows	RandomForestClassifier	Handles mixed features well
> 50,000 rows	GradientBoostingClassifier	Best accuracy on large data

Example

python

model = Classifier(data="titanic.csv", target="Survived")
model.train()                           # auto-select
model.train(algorithm="random_forest")  # explicit