KarlNaumann / MacroStat / 17580912954

97%

Build Type

push

github

Committed by web-flow

Commit Message

Autograd: gradient passthrough (#48)

* fix: add missing timestep to core forward

* feat: function to return a trainable instance of model behavior

* fix: simplify the self.history list update

* feat: timeseries as a list of torch.tensors, with .gather_timeseries to stack them. This propagates the gradient to the final timeseries

* fix: remove args from initialize_tensors

* fix: separate timeseries and timeseries_list, use gather_timeseries where appropriate

* test: update the variables test to adapt to the timeseries_list and timeseries differentiation

* ci: switch to only Python 3.13

* ci: update requirements file

* ci: add dash to requirements

* ci: workflow_dispatch trigger

* ci: match requirements.txt with setup.cfg

Run Details

239 of 241 branches covered (99.17%)

Branch coverage included in aggregate %.

41 of 50 new or added lines in 3 files covered. (82.0%)

4 existing lines in 1 file now uncovered.

1561 of 1620 relevant lines covered (96.36%)

0.96 hits per line