pystruct.models.
ChainCRF
(n_states=None, n_features=None, inference_method=None, class_weight=None, directed=True)[source]¶Linearchain CRF.
Pairwise potentials are symmetric and the same for all edges.
This leads to n_classes
parameters for unary potentials.
If directed=True
, there are n_classes * n_classes
parameters
for pairwise potentials, if directed=False
, there are only
n_classes * (n_classes + 1) / 2
(for a symmetric matrix).
Unary evidence x
is given as array of shape (n_nodes, n_features), and
labels y
are given as array of shape (n_nodes,). Chain lengths do not
need to be constant over the dataset.
Parameters:  n_states : int, default=None
inference_method : string or None, default=None
class_weight : None, or arraylike
directed : boolean, default=False


Methods
batch_inference (X, w[, relaxed]) 

batch_joint_feature (X, Y[, Y_true]) 

batch_loss (Y, Y_hat) 

batch_loss_augmented_inference (X, Y, w[, ...]) 

continuous_loss (y, y_hat) 

inference (x, w[, relaxed, return_energy]) 
Inference for x using parameters w. 
initialize (X, Y) 

joint_feature (x, y) 
Feature vector associated with instance (x, y). 
loss (y, y_hat) 

loss_augmented_inference (x, y, w[, relaxed, ...]) 
Lossaugmented Inference for x relative to y using parameters w. 
max_loss (y) 
__init__
(n_states=None, n_features=None, inference_method=None, class_weight=None, directed=True)[source]¶inference
(x, w, relaxed=False, return_energy=False)¶Inference for x using parameters w.
Finds (approximately) armin_y np.dot(w, joint_feature(x, y)) using self.inference_method.
Parameters:  x : tuple
w : ndarray, shape=(size_joint_feature,)
relaxed : bool, default=False
return_energy : bool, default=False


Returns:  y_pred : ndarray or tuple

joint_feature
(x, y)¶Feature vector associated with instance (x, y).
Feature representation joint_feature, such that the energy of the configuration (x, y) and a weight vector w is given by np.dot(w, joint_feature(x, y)).
Parameters:  x : tuple
y : ndarray or tuple


Returns:  p : ndarray, shape (size_joint_feature,)

loss_augmented_inference
(x, y, w, relaxed=False, return_energy=False)¶Lossaugmented Inference for x relative to y using parameters w.
Finds (approximately) armin_y_hat np.dot(w, joint_feature(x, y_hat)) + loss(y, y_hat) using self.inference_method.
Parameters:  x : tuple
y : ndarray, shape (n_nodes,)
w : ndarray, shape=(size_joint_feature,)
relaxed : bool, default=False
return_energy : bool, default=False


Returns:  y_pred : ndarray or tuple
