KowerKoint
KowerKoint
When a function is overloaded, documentation comments is placed in wrong place. For example, the C++ wrapper is below: ```cpp #include #include int add_int(int x, int y) { return x...
Difference between behavior of `add_state()` for `QuantumState` and `add_state()` for `DensityMatrix` cause misunderstanding. It is good to implement `make_superposition()` which makes a superposition `QuantumState` from two `QuantumState`s and `make_mixture()` which...
in gpusim, we use fixed block-size (ex: `1024` for most of single-qubit and double-qubit gate-apply, `512` for triple-qubit and quad-qubit dense-matrix-gate-aplly). In most of GPU, the fixed-value efficiently works, but...
close #625 I don't know why old dev insert `min(, 512)`, but this works with nqubits=25 in my environment. ```py from qulacs import QuantumStateGpu state = QuantumStateGpu(25) state.set_Haar_random_state() print(state.get_marginal_probability([1] *...
In https://github.com/qulacs/qulacs/blob/98a2751666dcdcc787df5db7e6e739d77a6ff1f8/src/gpusim/memory_ops.cu#L147 , maximum blocksize x gridsize = 512 x 512 = 2^18. So, state[2**18:] is not initialized correctly. ```py from qulacs import QuantumStateGpu state = QuantumStateGpu(19) state.set_Haar_random_state() v =...
get projection of specified control qubit and apply function to the part.
close #618 As I mentioned in the Issue, using `cudaOccupancyMaxPotentialBlockSize` to determine block size does not lost a performance and resolve errors because of insufficient registers in H100. In addition,...
Resolve error which occurs during upgrating swiftlint. I think this is responsible to swiftlint, but anyway `brew upgrade` is not needed.