add bypass metric (#1156)

* add bypass metric

* fixed `bypass` metric.

* add task attributes if predict_only

* add `predict_only` checks

* add docs

* added `overide_metric`, `override_config` to `Task`

* nits

* nit

* changed --predict_only to generations; nits

* nits

* nits

* change gen_kwargs warning

* add note about `--predict_only` in README.md

* added `predict_only`

* move table to bottom

* nit

* change null aggregation to bypass (conflict)

* bugfix; default `temp=0.0`

* typo

README.md

@@ -45,27 +45,7 @@ git clone https://github.com/EleutherAI/lm-evaluation-harness
 cd lm-evaluation-harness
 pip install -e .
 ```
-
-We also provide a number of optional dependencies for extended functionality. Extras can be installed via `pip install -e ".[NAME]"`
-
-| Name          | Use                                   |
-|---------------|---------------------------------------|
-| anthropic     | For using Anthropic's models          |
-| dev           | For linting PRs and contributions     |
-| gptq          | For loading models with GPTQ          |
-| ifeval        | For running the IFEval task           |
-| mamba         | For loading Mamba SSM models          |
-| math          | For running math task answer checking |
-| multilingual  | For multilingual tokenizers           |
-| openai        | For using OpenAI's models             |
-| optimum       | For running Intel OpenVINO models     |
-| promptsource  | For using PromptSource prompts        |
-| sentencepiece | For using the sentencepiece tokenizer |
-| testing       | For running library test suite        |
-| vllm          | For loading models with vLLM          |
-| zeno          | For visualizing results with Zeno     |
-|---------------|---------------------------------------|
-| all           | Loads all extras (not recommended)    |
+We also provide a number of optional dependencies for extended functionality. A detailed table is available at the end of this document.
 
 ## Basic Usage
 
@@ -204,6 +184,8 @@ A number of other libraries contain scripts for calling the eval harness through
 To create your own custom integration you can follow instructions from [this tutorial](https://github.com/EleutherAI/lm-evaluation-harness/blob/main/docs/interface.md#external-library-usage).
 
 ### Additional Features
+> [!Note]
+> For tasks unsuitable for direct evaluation — either due risks associated with executing untrusted code or complexities in the evaluation process — the `--predict_only` flag is available to obtain decoded generations for post-hoc evaluation.
 
 If you have a Metal compatible Mac, you can run the eval harness using the MPS back-end by replacing `--device cuda:0` with `--device mps` (requires PyTorch version 2.1 or higher).
 
@@ -254,7 +236,7 @@ Additionally, one can provide a directory with `--use_cache` to cache the result
 For a full list of supported arguments, check out the [interface](https://github.com/EleutherAI/lm-evaluation-harness/blob/main/docs/interface.md) guide in our documentation!
 
 > [!Tip]
-> Running lm-evaluation-harness as an external library and can't find (almost) any tasks available? run `lm_eval.tasks.initialize_tasks()` to load the library's stock tasks before calling `lm_eval.evaluate()` or `lm_eval.simple_evaluate()` !
+> Running lm-evaluation-harness as an external library and can't find (almost) any tasks available? Run `lm_eval.tasks.initialize_tasks()` to load the library's stock tasks before calling `lm_eval.evaluate()` or `lm_eval.simple_evaluate()` !
 
 ## Visualizing Results
 
@@ -319,6 +301,28 @@ We try to prioritize agreement with the procedures used by other groups to decre
 
 The best way to get support is to open an issue on this repo or join the [EleutherAI Discord server](https://discord.gg/eleutherai). The `#lm-thunderdome` channel is dedicated to developing this project and the `#release-discussion` channel is for receiving support for our releases. If you've used the library and have had a positive (or negative) experience, we'd love to hear from you!
 
+## Optional Extras
+Extras dependencies can be installed via `pip install -e ".[NAME]"`
+
+| Name          | Use                                   |
+|---------------|---------------------------------------|
+| anthropic     | For using Anthropic's models          |
+| dev           | For linting PRs and contributions     |
+| gptq          | For loading models with GPTQ          |
+| ifeval        | For running the IFEval task           |
+| mamba         | For loading Mamba SSM models          |
+| math          | For running math task answer checking |
+| multilingual  | For multilingual tokenizers           |
+| openai        | For using OpenAI's models             |
+| optimum       | For running Intel OpenVINO models     |
+| promptsource  | For using PromptSource prompts        |
+| sentencepiece | For using the sentencepiece tokenizer |
+| testing       | For running library test suite        |
+| vllm          | For loading models with vLLM          |
+| zeno          | For visualizing results with Zeno     |
+|---------------|---------------------------------------|
+| all           | Loads all extras (not recommended)    |
+
 ## Cite as
 
 ```

docs/interface.md

@@ -44,6 +44,8 @@ This mode supports a number of command-line arguments, the details of which can
 
 * `--include_path` : Accepts a path to a folder. If passed, then all YAML files containing `lm-eval`` compatible task configurations will be added to the task registry as available tasks. Used for when one is writing config files for their own task in a folder other than `lm_eval/tasks/`
 
+* `--predict_only`: Generates the model outputs without computing metrics. Use with `--log_samples` to retrieve decoded results.
+
 ## External Library Usage
 
 We also support using the library's external API for use within model training loops or other scripts.

lm_eval/__main__.py

@@ -143,6 +143,13 @@ def parse_eval_args() -> argparse.Namespace:
         metavar="CRITICAL|ERROR|WARNING|INFO|DEBUG",
         help="Controls the reported logging error level. Set to DEBUG when testing + adding new task configurations for comprehensive log output.",
     )
+    parser.add_argument(
+        "--predict_only",
+        "-x",
+        action="store_true",
+        default=False,
+        help="Use with --log_samples. Only model outputs will be saved and metrics will not be evaluated.",
+    )
     return parser.parse_args()
 
 
@@ -156,6 +163,11 @@ def cli_evaluate(args: Union[argparse.Namespace, None] = None) -> None:
     eval_logger.info(f"Verbosity set to {args.verbosity}")
     os.environ["TOKENIZERS_PARALLELISM"] = "false"
 
+    if args.predict_only:
+        args.log_samples = True
+    if (args.log_samples or args.predict_only) and not args.output_path:
+        assert args.output_path, "Specify --output_path"
+
     initialize_tasks(args.verbosity)
 
     if args.limit:
@@ -223,8 +235,6 @@ def cli_evaluate(args: Union[argparse.Namespace, None] = None) -> None:
         else:
             path.mkdir(parents=True, exist_ok=True)
             output_path_file = path.joinpath("results.json")
-    elif args.log_samples and not args.output_path:
-        assert args.output_path, "Specify --output_path"
 
     eval_logger.info(f"Selected Tasks: {task_names}")
 
@@ -243,6 +253,7 @@ def cli_evaluate(args: Union[argparse.Namespace, None] = None) -> None:
         write_out=args.write_out,
         log_samples=args.log_samples,
         gen_kwargs=args.gen_kwargs,
+        predict_only=args.predict_only,
     )
 
     if results is not None:

lm_eval/api/metrics.py

@@ -15,6 +15,11 @@ eval_logger = logging.getLogger("lm-eval")
 
 
 # Register Aggregations First
+@register_aggregation("bypass")
+def bypass_agg(arr):
+    return 999
+
+
 @register_aggregation("mean")
 def mean(arr):
     return sum(arr) / len(arr)
@@ -207,6 +212,16 @@ def mean_stderr(arr):
     return sample_stddev(arr) / math.sqrt(len(arr))
 
 
+@register_metric(
+    metric="bypass",
+    higher_is_better=True,
+    output_type=["loglikelihood", "multiple_choice", "generate_until"],
+    aggregation="bypass",
+)
+def bypass(items):
+    return None
+
+
 @register_metric(
     metric="mcc",
     higher_is_better=True,

lm_eval/api/task.py

@@ -1213,12 +1213,46 @@ class ConfigurableTask(Task):
 
         return result_dict
 
-    def aggregation(self):
+    def aggregation(self) -> dict:
         return self._aggregation_list
 
-    def higher_is_better(self):
+    def higher_is_better(self) -> dict:
         return self._higher_is_better
 
+    def get_config(self, key: str) -> Any:
+        return getattr(self._config, key, None)
+
+    def override_metric(self, metric_name: str) -> None:
+        """
+        Override the default metrics used for evaluation with custom metrics.
+
+        Parameters:
+        - metric_name (str): The name of the custom metric to override. Should be registered in api.metrics.
+        """
+        (
+            self._metric_fn_list,
+            self._aggregation_list,
+            self._metric_fn_kwargs,
+            self._higher_is_better,
+        ) = ({}, {}, {}, {})
+        self._metric_fn_list[metric_name] = get_metric(metric_name)
+        self._aggregation_list[metric_name] = get_metric_aggregation(metric_name)
+        self._higher_is_better[metric_name] = is_higher_better(metric_name)
+        self._metric_fn_kwargs[metric_name] = {}
+        setattr(self._config, "metric_list", [{"metric": metric_name}])
+        setattr(self._config, "process_results", None)
+
+    def override_config(
+        self, key: str = None, value: Any = None, update: bool = False
+    ) -> None:
+        if update:
+            current_value = getattr(self._config, key)
+            assert isinstance(current_value, dict)
+            current_value.update(value)
+            setattr(self._config, key, current_value)
+        else:
+            setattr(self._config, key, value)
+
 
 class MultipleChoiceTask(Task):
     OUTPUT_TYPE: str = "loglikelihood"

lm_eval/evaluator.py

@@ -38,6 +38,7 @@ def simple_evaluate(
     write_out: bool = False,
     log_samples: bool = True,
     gen_kwargs: str = None,
+    predict_only: bool = False,
 ):
     """Instantiate and evaluate a model on a list of tasks.
 
@@ -71,6 +72,9 @@ def simple_evaluate(
     :param gen_kwargs: str
         String arguments for model generation
         Ignored for all tasks with loglikelihood output_type
+    :param predict_only: bool
+        If true only model outputs will be generated and returned. Metrics will not be evaluated
+
     :return
         Dictionary of results
     """
@@ -89,7 +93,7 @@ def simple_evaluate(
     if gen_kwargs is not None:
         gen_kwargs = simple_parse_args_string(gen_kwargs)
         eval_logger.warning(
-            "generation_kwargs specified through cli, these settings will be used over set parameters in yaml tasks."
+            "generation_kwargs specified through cli, these settings will update set parameters in yaml tasks. Ensure 'do_sample=True' for non-greedy decoding!"
         )
         if gen_kwargs == "":
             gen_kwargs = None
@@ -129,25 +133,30 @@ def simple_evaluate(
             if task_obj is None:
                 continue
 
-        config = task_obj._config
-        if config["output_type"] == "generate_until" and gen_kwargs is not None:
-            config["generation_kwargs"].update(gen_kwargs)
+        if task_obj.get_config("output_type") == "generate_until":
+            if gen_kwargs is not None:
+                task_obj.override_config(
+                    key="generation_kwargs", value=gen_kwargs, update=True
+                )
+
+            if predict_only:
+                log_samples = True
+                eval_logger.info(
+                    f"Processing {task_name} in output-only mode. Metrics will not be calculated!"
+                )
+                # we have to change the class properties post-hoc. This is pretty hacky.
+                task_obj.override_metric(metric_name="bypass")
 
         if num_fewshot is not None:
-            if config["num_fewshot"] == 0:
+            if (default_num_fewshot := task_obj.get_config("num_fewshot")) == 0:
                 eval_logger.info(
                     f"num_fewshot has been set to 0 for {task_name} in its config. Manual configuration will be ignored."
                 )
             else:
-                default_num_fewshot = config["num_fewshot"]
-                if default_num_fewshot:
-                    # warn a user, if a specific num_fewshot > 0 was specified.
-                    # if unspecified in config, no warning message
-                    eval_logger.warning(
-                        f"Overwriting default num_fewshot of {task_name} from {default_num_fewshot} to {num_fewshot}"
-                    )
-
-                task_obj._config["num_fewshot"] = num_fewshot
+                eval_logger.warning(
+                    f"Overwriting default num_fewshot of {task_name} from {default_num_fewshot} to {num_fewshot}"
+                )
+                task_obj.override_config(key="num_fewshot", value=num_fewshot)
 
     if check_integrity:
         run_task_tests(task_list=tasks)
@@ -223,6 +232,14 @@ def evaluate(
 
     # decontaminate = decontamination_ngrams_path is not None
 
+    for task_name, task in task_dict.items():
+        if isinstance(task, tuple):
+            _, task = task
+        if not log_samples:
+            assert (
+                "bypass" not in getattr(task, "_metric_fn_list", {}).keys()
+            ), f"log_samples must be True for 'bypass' only tasks: {task_name}"
+
     # stores the final result for each task, for each metric/filter pair.
     results = collections.defaultdict(dict)
     # Tracks each task's version.
@@ -479,7 +496,8 @@ def evaluate(
         if bool(results):
             for group, task_list in reversed(task_hierarchy.items()):
                 if task_list == []:
-                    total_size = results[group]["samples"]
+                    # TODO: No samples when bypass
+                    total_size = results[group].get("samples", 999)
                 else:
                     total_size = 0
 

lm_eval/models/huggingface.py

@@ -715,12 +715,14 @@ class HFLM(LM):
                 return self.model(inps).logits
 
     def _model_generate(self, context, max_length, stop, **generation_kwargs):
+        # temperature = 0.0 if not set
         # if do_sample is false and temp==0.0:
         # remove temperature, as do_sample=False takes care of this
         # and we don't want a warning from HF
+        generation_kwargs["temperature"] = generation_kwargs.get("temperature", 0.0)
         do_sample = generation_kwargs.get("do_sample", None)
-        if do_sample is False and "temperature" == 0.0:
-            generation_kwargs.pop("temperature", 0.0)
+        if do_sample is False and generation_kwargs.get("temperature") == 0.0:
+            generation_kwargs.pop("temperature")
         # build stopping criteria
         stopping_criteria = stop_sequences_criteria(
             self.tokenizer, stop, context.shape[1], context.shape[0]