Neural network based solvers for partial differential equations and inverse problems :milky_way:. Implementation of physics-informed neural networks in pytorch.
In the current setup every n time-step a checkpoint was written, without overwriting the previous one. Which costs a lot of memory. However, the old checkpoints are in the most cases not considered anymore.
Update of the 1D schrodinger
The 1D schrodinger example was outdated and will be updated with this pull request.
Update in checkpointing
In the current setup every n time-step a checkpoint was written, without overwriting the previous one. Which costs a lot of memory. However, the old checkpoints are in the most cases not considered anymore.
Update of the 1D schrodinger
The 1D schrodinger example was outdated and will be updated with this pull request.