aadharsh06/pinn

This repository is a collection of physics-informed neural-network (PINN) implementations for learning PDE solutions, not a single application. The typical workflow is to choose a PDE formulation, train a neural network using initial/boundary-condition and physics-residual losses, then evaluate or plot predictions against reference solutions. The README identifies continuous-time, discrete-time, and coupled-equation examples, including Navier–Stokes in progress. Python training and evaluation sources were sampled for continuous-time examples; notebook-based branches are represented from the tree and README, so their internal wiring is not inferred.