QFlow 2.0: Quantum dot data for machine learning

Using a modified Thomas-Fermi approximation, we model a reference semiconductor system comprising a quasi-1D nanowire with a series of five depletion gates whose voltages determine the number of quantum dots (QDs), the charges on each of the QDs, as well as the conductance through the wire. The original dataset, QFlow lite, consists of 1 001 idealized simulated measurements with gate configurations sampling over different realizations of the same type of device. Each sample data is stored as a 100 x 100-pixel map from plunger gate voltages to (i) current through the device at infinitesimal bias, (ii) output of the charge sensor evaluated as the Coulomb potential at the sensor location - the experimentally relevant parameters that can be measured, (iii) information about the number of charges on each dot (with a default value 0 for short circuit and a barrier), and (iv) a label determining the state of the device, distinguishing between a single dot, a double dot, a short circuit, and a barrier state. The expanded dataset, QFlow 2.0, consists of 1599 idealized simulated measurements stored as roughly 250 x 250-pixel maps from plunger gate voltages to (i) output of the charge sensor, (ii) net charge on each dot, and (iii) a label determining the state of the device, distinguishing between a left, central, and right single QD, a double QD, and a barrier or short circuit (no QD) state. In addition, the QFlow 2.0 dataset includes two sets of noisy simulated measurements, one with the noise level varied around 1.5 times the optimized noise level and the other one with the noise level ranging from 0 to 7 times the optimized noise level. See the "Project description" and "Data structure" documents for additional information about these datasets.