Async eval callback (#702)
* Async eval callback
* add very basic tests
* more tests
* bump mcli
* woop, missing import
* instance not specified error
* fixes
* fix typing
* small testing fixes
* launch new run only on main process
* logger name
* items
* format
* Update llmfoundry/callbacks/async_eval_callback.py
Co-authored-by: Mihir Patel <mihir.v.patel7@gmail.com>
* Update llmfoundry/callbacks/async_eval_callback.py
Co-authored-by: Mihir Patel <mihir.v.patel7@gmail.com>
* feedback
* Apply suggestions from code review
Co-authored-by: Daniel King <43149077+dakinggg@users.noreply.github.com>
* small updates
* use parameters from train.py to capture overrides and mounted parameters file
* config_overrides
* updates
* fix test
* small fixes
* add logging
* remove last launch check
* better logging
* fix parameters
* fix double unit in the name
* sadz
* fies
* git integration path validation and update
* detect forks, better error/comment
* version import
* last checkpoint
* post_close -> close
* add todos, fix path bug
* add missing args
* remove eval_loader in callback too
* remove fit end event (already doing on close)
* misc fixes
* fix test
* add back eval interval
* build_loggers and add tests
* updates
* typing
* changes
* typing?
---------
Co-authored-by: Mihir Patel <mihir.v.patel7@gmail.com>
Co-authored-by: Daniel King <43149077+dakinggg@users.noreply.github.com>
