zyz111222 dd7e08c6db [Performance] Use forward_native for Conv3dLayer and add UT (#8375) ...

What this PR does / why we need it?
switch Ascend conv3d forward_oot to use forward_native and  add ut

Does this PR introduce any user-facing change?
No

How was this patch tested?
by CI

---------

Signed-off-by: zouyizhou <zouyizhou@huawei.com>

2026-04-20 17:20:40 +08:00

..

fla

[310P]: add torch chunk gated delta rule and 910b parity ut (#7594)

2026-03-25 16:46:43 +08:00

__init__.py

[Feature]: Support 310P device run qwen2.5/3 dense and qwen2.5vl models (#5776)

2026-01-17 11:49:18 +08:00

activation.py

[Feat.][310P]: weightNZ feature with quant or unquant. (#6705)

2026-02-13 15:41:02 +08:00

conv.py

[Performance] Use forward_native for Conv3dLayer and add UT (#8375)

2026-04-20 17:20:40 +08:00

layernorm.py

[310p]: add rmsnorm gated fallback and unit test (#7424)

2026-03-24 09:00:11 +08:00

mm_encoder_attention.py

[Bugfix][310p] the new A5 mmencoder op donot support 310p (#7518)

2026-03-23 15:40:34 +08:00

rotary_embedding.py

[Refact.]: Refactor some leftover implementations of 300I DUO in the main branch. (#6425)

2026-02-02 16:12:04 +08:00

vocab_parallel_embedding.py

[300I][Bugfix] fix unquant model weight nd2nz error (#6851)

2026-03-03 15:57:26 +08:00