dkamm
dkamm
- [x] refactor dfs to take in multiple input/output examples - [x] consider # nodes explored instead of wall time
- [x] Fix semantics of Variable - [x] iterate_inputs typemap is called with both a dict of type -> Variable list and a dict of type -> int index list....
- [ ] 2d embedding visualization - [ ] neural network prediction of example program - [ ] training summary (not provided in paper) - [ ] search speedups (dfs...
This seems like the correct thing to do? See https://beartype.readthedocs.io/en/latest/faq/#the-current-class and the following code sample ```python import ast from jaxtyping.import_hook import _JaxtypingTransformer transformer = _JaxtypingTransformer(typechecker="beartype.beartype") source = """ from typing_extensions...