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

[Misc] Split up pooling tasks (#10820)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2024-12-11 17:28:00 +08:00
committed by GitHub
parent 40766ca1b8
commit 8f10d5e393
27 changed files with 527 additions and 168 deletions
Download Patch File Download Diff File Expand all files Collapse all files

53
vllm/entrypoints/llm.py
View File

@@ -381,19 +381,20 @@ class LLM:
considered legacy and may be deprecated in the future. You should
instead pass them via the ``inputs`` parameter.
"""
task = self.llm_engine.model_config.task
if task != "generate":
runner_type = self.llm_engine.model_config.runner_type
if runner_type != "generate":
messages = [
"LLM.generate() is only supported for (conditional) generation "
"models (XForCausalLM, XForConditionalGeneration).",
]
supported_tasks = self.llm_engine.model_config.supported_tasks
if "generate" in supported_tasks:
supported_runner_types = self.llm_engine.model_config \
.supported_runner_types
if "generate" in supported_runner_types:
messages.append(
"Your model supports the 'generate' task, but is "
f"currently initialized for the '{task}' task. Please "
"initialize the model using `--task generate`.")
"Your model supports the 'generate' runner, but is "
f"currently initialized for the '{runner_type}' runner. "
"Please initialize vLLM using `--task generate`.")
raise ValueError(" ".join(messages))
@@ -793,16 +794,18 @@ class LLM:
considered legacy and may be deprecated in the future. You should
instead pass them via the ``inputs`` parameter.
"""
task = self.llm_engine.model_config.task
if task != "embedding":
messages = ["LLM.encode() is only supported for embedding models."]
runner_type = self.llm_engine.model_config.runner_type
if runner_type != "pooling":
messages = ["LLM.encode() is only supported for pooling models."]
supported_tasks = self.llm_engine.model_config.supported_tasks
if "embedding" in supported_tasks:
supported_runner_types = self.llm_engine.model_config \
.supported_runner_types
if "pooling" in supported_runner_types:
messages.append(
"Your model supports the 'embedding' task, but is "
f"currently initialized for the '{task}' task. Please "
"initialize the model using `--task embedding`.")
"Your model supports the 'pooling' runner, but is "
f"currently initialized for the '{runner_type}' runner. "
"Please initialize vLLM using `--task embed`, "
"`--task classify`, `--task score` etc.")
raise ValueError(" ".join(messages))
@@ -864,21 +867,23 @@ class LLM:
A list of ``PoolingRequestOutput`` objects containing the
generated scores in the same order as the input prompts.
"""
task = self.llm_engine.model_config.task
if task != "embedding":
messages = ["LLM.score() is only supported for embedding models."]
runner_type = self.llm_engine.model_config.runner_type
if runner_type != "pooling":
messages = ["LLM.score() is only supported for pooling models."]
supported_tasks = self.llm_engine.model_config.supported_tasks
if "embedding" in supported_tasks:
supported_runner_types = self.llm_engine.model_config \
.supported_runner_types
if "pooling" in supported_runner_types:
messages.append(
"Your model supports the 'embedding' task, but is "
f"currently initialized for the '{task}' task. Please "
"initialize the model using `--task embedding`.")
"Your model supports the 'pooling' runner, but is "
f"currently initialized for the '{runner_type}' runner. "
"Please initialize vLLM using `--task embed`, "
"`--task classify`, `--task score` etc.")
raise ValueError(" ".join(messages))
if not self.llm_engine.model_config.is_cross_encoder:
raise ValueError("Your model does not support the cross encoding")
raise ValueError("Your model does not support cross encoding")
tokenizer = self.llm_engine.get_tokenizer()