[SGLANG] Add the SGLANG generate API (#2997)

* add `sglang-generate`

* nit

* nit

* nit

* pacify pre-commit

lm_eval/api/task.py

@@ -401,7 +401,7 @@ class Task(abc.ABC):
         fewshot_as_multiturn: bool = False,
         chat_template: Optional[Callable] = None,
         tokenizer_name: str = "",
-        question_suffix: str = ""
+        question_suffix: str = "",
     ) -> None:
         """Build a set of Instances for a task, and store them in task.instances"""
 
@@ -1077,13 +1077,23 @@ class ConfigurableTask(Task):
         if not fewshot_as_multiturn:
             # if no messages or last message is system, append as new user entry
             if len(labeled_examples) == 0 or labeled_examples[-1]["role"] == "system":
-                labeled_examples.append({"role": "user", "content": question + question_suffix} if question_suffix else {"role": "user", "content": question}  )
+                labeled_examples.append(
+                    {"role": "user", "content": question + question_suffix}
+                    if question_suffix
+                    else {"role": "user", "content": question}
+                )
             # if last message is user, append to it to avoid two user messages in a row
             else:
-                labeled_examples[-1]["content"] += question + question_suffix if question_suffix else question
+                labeled_examples[-1]["content"] += (
+                    question + question_suffix if question_suffix else question
+                )
         else:
             # if fewshot_as_multiturn is True, append as next user entry (last is always assistant)
-            labeled_examples.append({"role": "user", "content": question + question_suffix} if question_suffix else {"role": "user", "content": question}  )
+            labeled_examples.append(
+                {"role": "user", "content": question + question_suffix}
+                if question_suffix
+                else {"role": "user", "content": question}
+            )
         if gen_prefix:
             labeled_examples.append({"role": "assistant", "content": gen_prefix})
 

lm_eval/models/__init__.py

@@ -16,6 +16,7 @@ from . import (
     optimum_ipex,
     optimum_lm,
     sglang_causallms,
+    sglang_generate_API,
     textsynth,
     vllm_causallms,
     vllm_vlms,

lm_eval/models/sglang_generate_API.py

+from typing import Dict, List, Optional, Tuple, Union
+
+from lm_eval.api.registry import register_model
+from lm_eval.models.openai_completions import LocalCompletionsAPI
+from lm_eval.models.utils import handle_stop_sequences
+
+
+@register_model("sglang-generate")
+class SGLANGGENERATEAPI(LocalCompletionsAPI):
+    def __init__(
+        self,
+        base_url=None,
+        tokenizer_backend="huggingface",
+        **kwargs,
+    ):
+        super().__init__(
+            base_url=base_url, tokenizer_backend=tokenizer_backend, **kwargs
+        )
+
+    def _create_payload(
+        self,
+        messages: Union[List[List[int]], List[dict], List[str], str],
+        generate=False,
+        gen_kwargs: Optional[dict] = None,
+        seed: int = 1234,
+        eos=None,
+        **kwargs,
+    ) -> dict:
+        is_string = (
+            True
+            if (isinstance(messages, str) or isinstance(messages[0], str))
+            else False
+        )
+        if generate:
+            gen_kwargs.pop("do_sample", False)
+            if "max_tokens" in gen_kwargs:
+                max_tokens = gen_kwargs.pop("max_tokens")
+            else:
+                max_tokens = gen_kwargs.pop("max_gen_toks", self._max_gen_toks)
+            temperature = gen_kwargs.pop("temperature", 0)
+            stop = handle_stop_sequences(gen_kwargs.pop("until", None), eos)
+            request = {
+                "sampling_params": {
+                    "max_new_tokens": max_tokens,
+                    "temperature": temperature,
+                    "stop": stop,
+                    **gen_kwargs,
+                },
+            }
+            request.update({"text": messages}) if is_string else request.update(
+                {"input_ids": messages}
+            )
+            return request
+        else:
+            assert not is_string, "Logprobs are only supported for tokenized inputs"
+            request = {
+                "input_ids": messages,
+                "sampling_params": {"max_new_tokens": 1, "temperature": 0},
+                "logprob_start_len": 0,
+                "top_logprobs_num": 1,
+                "return_logprob": True,
+            }
+            return request
+
+    @staticmethod
+    def parse_logprobs(
+        outputs: Union[Dict, List[Dict]],
+        tokens: List[List[int]] = None,
+        ctxlens: List[int] = None,
+        **kwargs,
+    ) -> List[Tuple[float, bool]]:
+        res = []
+        if not isinstance(outputs, list):
+            outputs = [outputs]
+        for choice, ctxlen in zip(outputs, ctxlens):
+            choice = choice["meta_info"]
+            assert ctxlen > 0, "Context length must be greater than 0"
+            logprobs = sum(x[0] for x in choice["input_token_logprobs"][ctxlen:])
+            is_greedy = all(
+                x[1] != y[0][1]
+                for x, y in zip(
+                    choice["input_token_logprobs"][ctxlen:],
+                    choice["input_top_logprobs"][ctxlen:],
+                )
+            )
+            res.append((logprobs, is_greedy))
+        return res
+
+    @staticmethod
+    def parse_generations(outputs: Union[Dict, List[Dict]], **kwargs) -> List[str]:
+        res = []
+        if not isinstance(outputs, list):
+            outputs = [outputs]
+        for out in outputs:
+            res.append(out["text"])
+        return res
+
+    @property
+    def api_key(self):
+        return ""

lm_eval/models/vllm_causallms.py

@@ -487,7 +487,10 @@ class VLLM(TemplateLM):
             inputs = []
             ctxlens = []
             for cache_key, context_enc, continuation_enc in chunk:
-                if full_length := len(context_enc + continuation_enc) >= self.max_length:
+                if (
+                    full_length := len(context_enc + continuation_enc)
+                    >= self.max_length
+                ):
                     eval_logger.warning(
                         f"Context length {full_length} exceeds max length ({self.max_length}). Truncating context."
                     )

lm_eval/tasks/c4/README.md

@@ -14,13 +14,13 @@ This is the processed version of Google's C4 dataset.
 
 ```text
 @misc{raffel2023exploringlimitstransferlearning,
-      title={Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer}, 
+      title={Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer},
       author={Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu},
       year={2023},
       eprint={1910.10683},
       archivePrefix={arXiv},
       primaryClass={cs.LG},
-      url={https://arxiv.org/abs/1910.10683}, 
+      url={https://arxiv.org/abs/1910.10683},
 }
 ```
 

lm_eval/tasks/c4/c4.yaml

@@ -21,4 +21,4 @@ dataset_kwargs:
     validation: en/c4-validation.00000-of-00008.json.gz
   # following the choice of https://arxiv.org/abs/2410.07461
   trust_remote_code: true
-  verification_mode: "no_checks"
\ No newline at end of file
+  verification_mode: "no_checks"

lm_eval/tasks/mmlu_pro/utils.py

 from functools import partial
 
+
 choices = ["A", "B", "C", "D", "E", "F", "G", "H", "I", "J"]
 
+
 def format_cot_example(example, including_answer=True):
     prompt = "Question:\n"
     question = example["question"]
@@ -21,15 +23,18 @@ def format_cot_example(example, including_answer=True):
         prompt += cot_content + "\n\n"
     else:
         prompt += "Answer: Let's think step by step."
-    
+
     return prompt
 
+
 doc_to_text = partial(format_cot_example, including_answer=False)
 fewshot_to_text = partial(format_cot_example, including_answer=True)
 
+
 def process_docs(dataset, subject):
     return dataset.filter(lambda x: x["category"] == subject)
 
+
 process_biology = partial(process_docs, subject="biology")
 process_business = partial(process_docs, subject="business")
 process_chemistry = partial(process_docs, subject="chemistry")