biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Refactor] Separate sequence and token pooling types (#32026)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2026-01-10 12:53:24 +08:00
committed by GitHub
parent 52d428295d
commit 583a90e005
42 changed files with 324 additions and 204 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tests/model_executor/test_model_load_with_params.py
View File

@@ -46,7 +46,8 @@ def test_model_loading_with_params(vllm_runner, monkeypatch):
assert model_config.encoder_config["do_lower_case"]
# asserts on the pooling config files
assert model_config.pooler_config.pooling_type == "CLS"
assert model_config.pooler_config.seq_pooling_type == "CLS"
assert model_config.pooler_config.tok_pooling_type == "ALL"
assert model_config.pooler_config.normalize
# asserts on the tokenizer loaded
@@ -90,7 +91,8 @@ def test_roberta_model_loading_with_params(vllm_runner, monkeypatch):
assert not model_config.encoder_config["do_lower_case"]
# asserts on the pooling config files
assert model_config.pooler_config.pooling_type == "MEAN"
assert model_config.pooler_config.seq_pooling_type == "MEAN"
assert model_config.pooler_config.tok_pooling_type == "ALL"
assert model_config.pooler_config.normalize
# asserts on the tokenizer loaded