George Hotz 1b4618e257 gradient cleanup (#8750) ...

* switch backward to use gradient [pr]

* set device correctly, dedup

* why does that fail?

* add noop cast

* simple backward

* fix beautiful_mnist

* touchups

* set in compute_gradient

* uop_count

* uop_count was wrong

* collections

* no note

* skip that test

* update sched kernel counts

* train mnist is 65

* fix metadata and gc

* fixes

* materialize_grads

* no pathlib stuff

* add contiguous_backward, fix bugs

* add some realize

* fix multi

* remove unused backward passes [pr]

* lower line count

2025-01-26 09:30:55 +09:00

..

benchmark.yml

add bert to bechmark ci (#8741)

2025-01-24 14:45:11 -05:00

docs.yml

add strict mkdocs check (#5497)

2024-07-15 14:21:37 -07:00

python-publish.yml

update gh actions (#3033)

2024-01-09 17:52:22 -08:00

szdiff.yml

update gh actions (#3033)

2024-01-09 17:52:22 -08:00

test.yml

gradient cleanup (#8750)

2025-01-26 09:30:55 +09:00