tick.simulation.weights_sparse_exp¶
-
tick.simulation.weights_sparse_exp(n_weigths: int = 100, nnz: int = 10, scale: float = 10.0, dtype='float64') → numpy.ndarray[source]¶ Sparse and exponential model weights generator
Instance of weights for a model, given by a vector with exponentially decaying components: the j-th entry is given by
for 0 <= j <= nnz - 1. For j >= nnz, the entry is zero.
- Parameters
n_weigths :
int, default=100Number of weights
nnz :
int, default=10Number of non-zero weights
scale :
float, default=10.The scaling of the exponential decay
dtype :
{'float64', 'float32'}, default=’float64’Type of the arrays used.
- Returns
output : np.ndarray, shape=(n_weigths,)
The weights vector
