QDriftSampler

class QDriftSampler(gate, num_terms, weights, rng)

Bases: object

The draws of the qDRIFT protocol for one Evolution gate.

This is what FermionicQDrift.sampler() returns. It holds the distribution derived from the gate, and exists so that a caller that needs to look at each draw (the QDriftTrotterization pass, when it rejects draws) can take them one at a time, whereas FermionicQDrift.synthesize() takes them all at once.

Initializing a sampler can be done with the arguments below.

Parameters:

• gate (Evolution) – the gate to sample.

• num_terms (int) – the number of gates the whole sample holds.

• weights (np.ndarray | None) – the sampling weights, or None to derive them from the operator of gate.

• rng (np.random.Generator) – the random number generator to draw from.

Raises:

ValueError – if weights does not hold one entry per group (or term) of the operator.

Attributes

time

The evolution time \(\delta\) of every sampled gate.

Methods

draw()

Draws a single gate.

Returns:

The drawn atomic Evolution gate.

Return type:

Evolution

sample(size=None)

Draws size gates in a single batched call.

Parameters:

size (int | None) – the number of gates to draw. Defaults to num_terms.

Returns:

The drawn atomic Evolution gates, in the order of the draws.

Return type:

list[Evolution]