Caching ``__dask_graph__``
Open
- Dominant language
- Python
- Stars
- 89
- Forks
- 26
- PR merge metrics
- No merged PRs in 30d
Description
Currently, we are building the graph multiple times before doing the actual computation. This wasn't a problem, since the optimizations were cheap. We should cache the graph to avoid multiple passthroughs as soon as we start adding methods that might trigger computations during the optimization step (see #181)
Contributor guide
Assessment
This issue has not been assessed yet.