Movatterモバイル変換

Skip to content

NVIDIA/TensorRT-LLMPublic

NotificationsYou must be signed in to change notification settings
Fork1.9k
Star12.3k

Commitcceb4f0

xinhe-nv

committed

add gc for test_llama_3_x_fp8_with_bf16_lora

Signed-off-by: Xin He (SW-GPU) <200704525+xinhe-nv@users.noreply.github.com>

1 parentd3059db commitcceb4f0Copy full SHA for cceb4f0

File tree

1 file changed

+2

-0

lines changed

tests/integration/defs
- conftest.py

1 file changed

+2

-0

lines changed

`‎tests/integration/defs/conftest.py‎`

Lines changed: 2 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -15,6 +15,7 @@`
`15`	`15`	`# -- coding: utf-8 --`
`16`	`16`
`17`	`17`	`importdatetime`
	`18`	`+importgc`
`18`	`19`	`importos`
`19`	`20`	`importplatform`
`20`	`21`	`importre`
`@@ -2524,6 +2525,7 @@ def torch_empty_cache() -> None:`
`2524`	`2525`	`"""`
`2525`	`2526`	`iftorch.cuda.is_available():`
`2526`	`2527`	`torch.cuda.empty_cache()`
	`2528`	`+gc.collect()`
`2527`	`2529`
`2528`	`2530`
`2529`	`2531`	`@pytest.fixture(autouse=True)`

0 commit comments

Comments

(0)

[8]ページ先頭

©2009-2025 Movatter.jp