Pedro Cuenca 67a00e26ed Run ControlNet compile test in a separate subprocess ...

`torch.compile()` spawns several subprocesses and the GPU memory used
was not reclaimed after the test ran. This approach was taken from
`transformers`.

2023-05-22 11:45:51 +02:00

..

fixtures

Fix config prints and save, load of pipelines (#2849)

2023-04-11 13:35:42 +02:00

models

parameterize pass single args through tuple (#3477)

2023-05-18 10:14:29 -07:00

others

[Tests] better determinism (#3374)

2023-05-11 16:38:14 +01:00

pipelines

Run ControlNet compile test in a separate subprocess

2023-05-22 11:45:51 +02:00

schedulers

mps & onnx tests rework (#3449)

2023-05-20 13:43:07 +02:00

__init__.py

fix issues with loading, add test for pipeline

2022-06-07 15:40:36 +02:00

conftest.py

[Copyright] 2023 (#2524)

2023-03-01 10:31:00 +01:00