Quantum circuit dataset

Dataset for quantum circuits.

Simple Dataset

source

CircuitsConfigDatasetConfig

 CircuitsConfigDatasetConfig (store_dict:dict, dataset_to_gpu:bool,
                              optimized:bool, random_samples:int,
                              num_of_qubits:int, min_gates:int,
                              max_gates:int, max_params:int,
                              gate_pool:list[str])

source

CircuitsConfigDataset

 CircuitsConfigDataset (device:torch.device=device(type='cpu'),
                        **parameters)

Dataset for quantum circuits, access gate_pool directly and all other paras with .params_config

init = {k:None for k in CircuitsConfigDataset.req_params}
init["gate_pool"]  = ["qiskit.circuit.library.standard_gates.h.HGate",
                      "qiskit.circuit.library.standard_gates.x.CXGate"]
init["store_dict"] = {"x":"tensor", "y":"tensor_list"}

a = CircuitsConfigDataset(**init)
a.get_config()
{'target': '__main__.CircuitsConfigDataset',
 'device': 'cpu',
 'comment': '',
 'save_path': None,
 'save_datetime': '06/01/2025 11:31:35',
 'save_type': 'safetensors',
 'params': CircuitsConfigDatasetConfig(store_dict={'x': 'tensor', 'y': 'tensor_list'}, dataset_to_gpu=None, optimized=None, random_samples=None, num_of_qubits=None, min_gates=None, max_gates=None, max_params=None, gate_pool=['qiskit.circuit.library.standard_gates.h.HGate', 'qiskit.circuit.library.standard_gates.x.CXGate'])}

Mixed Dataset

source

MixedCircuitsConfigDatasetConfig

 MixedCircuitsConfigDatasetConfig (store_dict:dict, dataset_to_gpu:bool,
                                   pad_constant:int, collate_fn:str,
                                   bucket_batch_size:int,
                                   model_scale_factor:int, optimized:bool,
                                   random_samples:int, num_of_qubits:int,
                                   min_gates:int, max_gates:int,
                                   max_params:int, gate_pool:list[str])

source

MixedCircuitsConfigDataset

 MixedCircuitsConfigDataset (device:torch.device=device(type='cpu'),
                             **parameters)

Dataset that uses multiple cached dataset and combines them with padding, either i) Bucket or ii) Max. Also provides a corresponding collate_fn for training.

Back to top