Our project predicts how long a quantum circuit will take to simulate, given the circuit itself and a simulation threshold. We built a fast pure-Python QASM parser that extracts a set of 50 meaningful circuit features while reading each circuit exactly once. These features feed a LightGBM-based model trained on log10 runtime, evaluated with grouped cross-validation so no circuit's data leaks between train and test folds, and validated on a held-out 20% split of circuits the model never saw during training or tuning. The result is a model that scores 92.86% against provided true runtimes for the held-out set.
The Wolfpack
Solution to the Quantum Rings challenge.
Updates
Leave feedback in the comments!
Log in or sign up for Devpost to join the conversation.