AAPred.fit

AAPred.fit(X, labels, label_pos=1, optimize_hyperparams=False, param_grids=None, n_cv=5)[source]

Fit every model on the full dataset for deployment.

Each model class from the constructor is instantiated and fit on all of X / labels; the fitted estimators are kept in list_models_ and reused by predict() and eval().

Added in version 1.1.0.

Parameters:

X (array-like, shape (n_samples, n_features)) – Feature matrix. Rows typically correspond to samples and columns to features.
labels (array-like, shape (n_samples,)) – Class labels for samples in X (typically 1 for the positive class and 0 for the negative class).
label_pos (int, default=1) – Label of the positive class whose probability predict() scores.
optimize_hyperparams (bool, default=False) – If True, each model is tuned by GridSearchCV (n_cv folds) over its param_grids entry, or a built-in default grid when none is given; the best estimator is kept. If False, models are fit with their given parameters.
param_grids (dict or list of dict, optional) – Hyperparameter grid(s) for the optimization. A single dict is applied to every model; a list must have one grid per model. Used only when optimize_hyperparams=True.
n_cv (int, default=5) – Number of stratified cross-validation folds used by the hyperparameter search.

Returns:

The fitted AAPred instance (self).

Return type:

AAPred

Examples

To demonstrate AAPred().fit(), we obtain the DOM_GSEC dataset and its feature matrix:

import aaanalysis as aa
aa.options["verbose"] = False  # Disable verbosity

# DOM_GSEC example dataset + its feature set (see [Breimann25]_)
df_seq = aa.load_dataset(name="DOM_GSEC")
labels = df_seq["label"].to_list()
df_feat = aa.load_features(name="DOM_GSEC").head(20)

# Build the CPP feature matrix
sf = aa.SequenceFeature()
df_parts = sf.get_df_parts(df_seq=df_seq)
X = sf.feature_matrix(features=df_feat["feature"], df_parts=df_parts)

Fitting trains every model on the full dataset and stores them for deployment in list_models_:

aapred = aa.AAPred(random_state=42)
aapred.fit(X, labels)
print("Number of fitted models:", len(aapred.list_models_))

Number of fitted models: 1

Multiple model classes can be evaluated and deployed together. Any scikit-learn classifier implementing predict_proba is accepted:

from sklearn.svm import SVC
aapred = aa.AAPred(list_model_classes=[SVC], list_model_kwargs=[{"probability": True, "kernel": "linear"}],
                   random_state=42)
aapred.fit(X, labels)
print("Number of fitted models:", len(aapred.list_models_))

Number of fitted models: 1

/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(

The positive class whose probability :meth:AAPred.predict_proba returns is set by label_pos (default=1).

Further parameters. label_pos sets which class is treated as positive. Hyperparameters can be tuned per model by GridSearchCV: enable optimize_hyperparams, optionally passing an explicit param_grids (a single dict is applied to every model) and the number of stratified folds n_cv:

aapred = aa.AAPred(models=["svm"], random_state=42)
aapred.fit(X, labels, label_pos=1, optimize_hyperparams=True,
           param_grids={"C": [0.1, 1.0, 10.0]}, n_cv=5)
print("Number of fitted models:", len(aapred.list_models_))

Number of fitted models: 1

/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(
/Users/stephanbreimann/Programming/1Packages/aaanalysis/.venv/lib/python3.13/site-packages/sklearn/svm/_base.py:239: FutureWarning: The probability parameter was deprecated in 1.9 and will be removed in version 1.11. Use CalibratedClassifierCV(SVC(), ensemble=False) instead of SVC(probability=True)
  warnings.warn(