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

[Bugfix] Use runner_type instead of task in GritLM (#11144)

Browse Source 
Signed-off-by: Pooya Davoodi <pooya.davoodi@parasail.io>
This commit is contained in:
Pooya Davoodi
2024-12-12 20:09:53 -08:00
committed by GitHub
parent 30870b4f66
commit 1efce68605
2 changed files with 7 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tests/models/embedding/language/test_gritlm.py
View File

@@ -35,7 +35,7 @@ def test_find_array(monkeypatch):
from vllm.model_executor.models.gritlm import GritLMPooler
# Create an LLM object to get the model config.
llm = vllm.LLM(MODEL_NAME, task="embedding", max_model_len=MAX_MODEL_LEN)
llm = vllm.LLM(MODEL_NAME, task="embed", max_model_len=MAX_MODEL_LEN)
pooler = GritLMPooler(model_config=llm.llm_engine.model_config)
arr = _arr([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])
@@ -55,7 +55,7 @@ def server_embedding():
with pytest.MonkeyPatch.context() as mp:
mp.setenv("VLLM_ATTENTION_BACKEND", "XFORMERS")
args = ["--task", "embedding", "--max_model_len", str(MAX_MODEL_LEN)]
args = ["--task", "embed", "--max_model_len", str(MAX_MODEL_LEN)]
with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:
yield remote_server
@@ -141,7 +141,7 @@ def test_gritlm_offline_embedding(monkeypatch):
queries, q_instruction, documents, d_instruction = get_test_data()
llm = vllm.LLM(MODEL_NAME, task="embedding", max_model_len=MAX_MODEL_LEN)
llm = vllm.LLM(MODEL_NAME, task="embed", max_model_len=MAX_MODEL_LEN)
d_rep = run_llm_encode(
llm,