"""
Generalized Matrix LVQ (GMLVQ).
GLVQ with a learned linear transformation :math:`\\Omega` that maps data into
a discriminative subspace:
.. math::
d(x, w) = \\|\\Omega(x - w)\\|^2
References
----------
.. [1] Schneider, P., Biehl, M., & Hammer, B. (2009). Adaptive
Relevance Matrices in Learning Vector Quantization. Neural
Computation.
"""
import jax
import jax.numpy as jnp
from jax import jit
import numpy as np
from prosemble.models.prototype_base import SupervisedPrototypeModel
from prosemble.core.competitions import wtac
from prosemble.core.initializers import identity_omega_init
@jit
def _predict_gmlvq_jit(X, prototypes, omega, proto_labels):
"""JIT-compiled GMLVQ prediction with learned :math:`\\Omega` metric."""
diff = X[:, None, :] - prototypes[None, :, :]
projected = jnp.einsum('npd,dl->npl', diff, omega)
distances = jnp.sum(projected ** 2, axis=2)
return wtac(distances, proto_labels)
[docs]
class GMLVQ(SupervisedPrototypeModel):
"""Generalized Matrix Learning Vector Quantization.
Learns a global linear mapping :math:`\\Omega` (d x latent_dim) such that
distances are computed in the transformed space:
.. math::
d(x, w) = (x - w)^T \\Omega^T \\Omega (x - w)
The relevance matrix :math:`\\Lambda = \\Omega^T \\Omega` captures feature
correlations.
Parameters
----------
latent_dim : int, optional
Dimensionality of the latent space. If None, uses input dim.
beta : float
Transfer function steepness.
n_prototypes_per_class : int
Number of prototypes per class.
max_iter : int
Maximum training iterations.
lr : float
Learning rate.
epsilon : float
Convergence threshold on loss change.
random_seed : int
Random seed for reproducibility.
distance_fn : callable, optional
Distance function (default: squared Euclidean).
optimizer : str or optax optimizer, optional
Optimizer name ('adam', 'sgd') or optax GradientTransformation.
Default: 'adam'.
transfer_fn : callable, optional
Transfer function for loss shaping (default: identity).
margin : float
Margin for loss computation.
callbacks : list, optional
List of Callback objects.
use_scan : bool
If True (default), use jax.lax.scan for training (faster, JIT-compiled,
but runs all max_iter iterations even after convergence).
If False, use a Python for-loop with true early stopping (no wasted
compute after convergence, but slower per iteration).
batch_size : int, optional
Mini-batch size. If None (default), use full-batch training.
When set, each epoch iterates over shuffled mini-batches of this size.
lr_scheduler : str or optax.Schedule, optional
Learning rate schedule. Supported strings: 'exponential_decay',
'cosine_decay', 'warmup_cosine_decay', 'warmup_exponential_decay',
'warmup_constant', 'polynomial', 'linear', 'piecewise_constant',
'sgdr'. Or pass a custom optax.Schedule. Default: None.
lr_scheduler_kwargs : dict, optional
Keyword arguments passed to the learning rate scheduler
(e.g. ``decay_rate``, ``transition_steps``). Default: None.
prototypes_initializer : str or callable, optional
How to initialize prototypes. Supported strings: 'stratified_random'
(default), 'class_mean', 'class_conditional_mean', 'stratified_noise',
'random_normal', 'uniform', 'zeros', 'ones', 'fill_value'.
Or pass a callable ``(X, y, n_per_class, key) -> (protos, labels)``.
patience : int, optional
Number of consecutive epochs with no improvement before stopping.
If None (default), stops after a single non-improving step (epsilon
check). Requires use_scan=False for true early stopping.
restore_best : bool
If True, restore the parameters that achieved the lowest loss
(or validation loss if validation data is provided). Default: False.
class_weight : dict or 'balanced', optional
Weights for each class. Dict maps class label to weight, e.g.
{0: 1.0, 1: 2.0, 2: 1.5}. 'balanced' auto-computes weights
inversely proportional to class frequencies. Default: None (uniform).
gradient_accumulation_steps : int, optional
Accumulate gradients over this many steps before applying an update.
Effective batch size = batch_size * gradient_accumulation_steps.
Default: None (no accumulation).
ema_decay : float, optional
Exponential moving average decay for parameters (0 < ema_decay < 1).
After training, model parameters are replaced with EMA-smoothed values.
Typical values: 0.999, 0.9999. Default: None (no EMA).
freeze_params : list of str, optional
List of parameter group names to freeze (zero gradients).
E.g. ['backbone'] to freeze the backbone and only train prototypes.
Default: None (all parameters trainable).
lookahead : dict, optional
Enable lookahead optimizer wrapper. Dict with keys:
- 'sync_period': int (default 6) — sync every k steps
- 'slow_step_size': float (default 0.5) — interpolation factor
Default: None (no lookahead).
mixed_precision : str or None, optional
Compute dtype for mixed precision training. 'float16' or 'bfloat16'.
Master weights stay in float32; forward/backward pass runs in lower
precision for ~2x speed and ~half memory on GPU. Float16 uses static
loss scaling to prevent gradient underflow. Default: None (disabled).
"""
def __init__(self, latent_dim=None, beta=10.0,
n_prototypes_per_class=1, max_iter=100, lr=0.01,
epsilon=1e-6, random_seed=42, distance_fn=None,
optimizer='adam', transfer_fn=None, margin=0.0,
callbacks=None, use_scan=True, batch_size=None,
lr_scheduler=None, lr_scheduler_kwargs=None,
prototypes_initializer=None, patience=None,
restore_best=False, class_weight=None,
gradient_accumulation_steps=None, ema_decay=None,
freeze_params=None, lookahead=None, mixed_precision=None):
super().__init__(
n_prototypes_per_class=n_prototypes_per_class,
max_iter=max_iter, lr=lr, epsilon=epsilon,
random_seed=random_seed, distance_fn=distance_fn,
optimizer=optimizer, transfer_fn=transfer_fn, margin=margin,
callbacks=callbacks, use_scan=use_scan, batch_size=batch_size,
lr_scheduler=lr_scheduler,
lr_scheduler_kwargs=lr_scheduler_kwargs,
prototypes_initializer=prototypes_initializer,
patience=patience, restore_best=restore_best,
class_weight=class_weight,
gradient_accumulation_steps=gradient_accumulation_steps,
ema_decay=ema_decay, freeze_params=freeze_params,
lookahead=lookahead, mixed_precision=mixed_precision,
)
self.latent_dim = latent_dim
self.beta = beta
self.omega_ = None
def _get_resume_params(self, params):
params['omega'] = self.omega_
return params
def _init_state(self, X, y, key):
n_features = X.shape[1]
latent_dim = self.latent_dim or n_features
key1, key2 = jax.random.split(key)
prototypes, proto_labels = self._init_prototypes(
X, y, self.n_prototypes_per_class, key1
)
omega = identity_omega_init(n_features, latent_dim)
params = {'prototypes': prototypes, 'omega': omega}
opt_state = self._optimizer.init(params)
from prosemble.models.prototype_base import SupervisedState
state = SupervisedState(
prototypes=prototypes,
opt_state=opt_state,
loss=jnp.array(float('inf')),
iteration=0,
converged=False,
)
return state, params, proto_labels
def _compute_loss(self, params, X, y, proto_labels):
prototypes = params['prototypes']
omega = params['omega']
# Omega distance: d(x, w) = ||Omega(x - w)||^2
diff = X[:, None, :] - prototypes[None, :, :] # (n, p, d)
projected = jnp.einsum('npd,dl->npl', diff, omega) # (n, p, l)
distances = jnp.sum(projected ** 2, axis=2) # (n, p)
from prosemble.core.losses import glvq_loss_with_transfer
return glvq_loss_with_transfer(
distances, y, proto_labels,
transfer_fn=self.transfer_fn,
margin=self.margin,
beta=self.beta,
)
def _compute_distances_for_rejection(self, X):
"""Omega-projected distances for reject option."""
diff = X[:, None, :] - self.prototypes_[None, :, :]
projected = jnp.einsum('npd,dl->npl', diff, self.omega_)
return jnp.sum(projected ** 2, axis=2)
def _extract_results(self, params, proto_labels, loss_history, n_iter, **kwargs):
super()._extract_results(params, proto_labels, loss_history, n_iter, **kwargs)
self.omega_ = params['omega']
@property
def omega_matrix(self):
"""Return the learned :math:`\\Omega` matrix."""
if self.omega_ is None:
raise ValueError("Model not fitted.")
return self.omega_
@property
def lambda_matrix(self):
"""Return :math:`\\Lambda = \\Omega^T \\Omega` (relevance matrix)."""
if self.omega_ is None:
raise ValueError("Model not fitted.")
return self.omega_.T @ self.omega_
[docs]
def predict(self, X):
"""Predict using learned :math:`\\Omega` distance."""
self._check_fitted()
X = jnp.asarray(X, dtype=jnp.float32)
return _predict_gmlvq_jit(
X, self.prototypes_, self.omega_, self.prototype_labels_
)
def _get_quantizable_attrs(self):
attrs = super()._get_quantizable_attrs()
if self.omega_ is not None:
attrs.append('omega_')
return attrs
def _get_fitted_arrays(self):
arrays = super()._get_fitted_arrays()
if self.omega_ is not None:
arrays['omega_'] = np.asarray(self.omega_)
return arrays
def _set_fitted_arrays(self, arrays):
super()._set_fitted_arrays(arrays)
if 'omega_' in arrays:
self.omega_ = jnp.asarray(arrays['omega_'])
def _get_hyperparams(self):
hp = super()._get_hyperparams()
hp['beta'] = self.beta
if self.latent_dim is not None:
hp['latent_dim'] = self.latent_dim
return hp