parcels

parcels copied to clipboard

Expand tests to cases for interaction kernels that are also expected to fail. The tests that are currently implemented mainly test scenarios that are expected to succeed (and not so...

CKehl

Create a good, illustrative example Jupyter notebook (in collaboration with @fedeguerr @pdnooteboom and @mikaelk ) demonstrating the interaction kernel usage, as well as including the following things: - execution order...

CKehl

Particle-particle interaction does currently not work with MPI. That is mainly because particles interacting with one another potentially or probably occupy different processors. Hence, they would find each other as...

CKehl

After having implemented Interaction kernels for JIT, implement this interaction in SciPy and JIT for AoS particle sets. (subtasks still to-be-determined)

CKehl

After having implemented Interaction kernels for JIT, implement this interaction in SciPy and JIT for Nodes particle sets. (subtasks still to-be-determined)

CKehl

extend InteractionKernel functionality by making it work in a Kernel-generated, C-jit environment. Benchmark first changes where only the kernel execution is integrated in JIT. If benchmark says neighbourhood parsing is...

CKehl

The execute loop of InteractionKernels currently does not handle State/Operation codes that signify an error or edge case. It only reports a warning. It does not use the State-/OperationCode system...

CKehl

Fix the issue of small dt - long runtime [to the degree feasible and possible], where dt of milliseconds and runtime of 10-years or more leads to time increment deviation...

CKehl

Index search on skewed grid

1

We're working with a somewhat skewed grid. For example, the x-coordinates at the x=0 index look like ``` array([113.466435, 113.469376, 113.472318, ..., 122.49621 , 122.499151, 122.502092]) ``` and similarly the...

angus-g

Better ways of storing particle locations for connectivity-related studies?

2

In many studies related to connectivity (e.g. ecological, marine debris source attribution, etc.) we are less interested in the specific trajectory of the particle and more interested in discrete events,...

nvogtvincent