Generate: add config-level validation (#25381)

src/transformers/generation/configuration_utils.py

@@ -332,12 +332,121 @@ class GenerationConfig(PushToHubMixin):
 
     def validate(self):
         """
-        Validates the values of the attributes of the GenerationConfig instance, and raises a `ValueError` if any of
+        Validates the values of the attributes of the [`GenerationConfig`] instance. Raises exceptions in the presence
-        the values are invalid.
+        of parameterization that can be detected as incorrect from the configuration instance alone.
+
+        Note that some parameters are best validated at generate runtime, as they may depend on other inputs and/or the
+        model, such as parameters related to the generation length.
         """
+
+        # Validation of individual attributes
         if self.early_stopping not in {True, False, "never"}:
             raise ValueError(f"`early_stopping` must be a boolean or 'never', but is {self.early_stopping}.")
 
+        # Validation of attribute relations:
+        # 1. detect sampling-only parameterization when not in sampling mode
+        if self.do_sample is False:
+            greedy_wrong_parameter_msg = (
+                "`do_sample` is set to `False`. However, {flag_name} is set to {flag_value} -- this flag is only used "
+                "in sample-based generation modes. Set `do_sample=True` or unset {flag_name} to continue."
+            )
+            if self.temperature != 1.0:
+                raise ValueError(
+                    greedy_wrong_parameter_msg.format(flag_name="temperature", flag_value=self.temperature)
+                )
+            if self.top_p != 1.0:
+                raise ValueError(greedy_wrong_parameter_msg.format(flag_name="top_p", flag_value=self.top_p))
+            if self.typical_p != 1.0:
+                raise ValueError(greedy_wrong_parameter_msg.format(flag_name="typical_p", flag_value=self.typical_p))
+            if self.top_k != 50 and self.penalty_alpha is None:  # contrastive search uses top_k
+                raise ValueError(greedy_wrong_parameter_msg.format(flag_name="top_k", flag_value=self.top_k))
+            if self.epsilon_cutoff != 0.0:
+                raise ValueError(
+                    greedy_wrong_parameter_msg.format(flag_name="epsilon_cutoff", flag_value=self.epsilon_cutoff)
+                )
+            if self.eta_cutoff != 0.0:
+                raise ValueError(greedy_wrong_parameter_msg.format(flag_name="eta_cutoff", flag_value=self.eta_cutoff))
+
+        # 2. detect beam-only parameterization when not in beam mode
+        if self.num_beams == 1:
+            single_beam_wrong_parameter_msg = (
+                "`num_beams` is set to 1. However, {flag_name} is set to {flag_value} -- this flag is only used in "
+                "beam-based generation modes. Set `num_beams>1` or unset {flag_name} to continue."
+            )
+            if self.early_stopping is not False:
+                raise ValueError(
+                    single_beam_wrong_parameter_msg.format(flag_name="early_stopping", flag_value=self.early_stopping)
+                )
+            if self.num_beam_groups != 1:
+                raise ValueError(
+                    single_beam_wrong_parameter_msg.format(
+                        flag_name="num_beam_groups", flag_value=self.num_beam_groups
+                    )
+                )
+            if self.diversity_penalty != 0.0:
+                raise ValueError(
+                    single_beam_wrong_parameter_msg.format(
+                        flag_name="diversity_penalty", flag_value=self.diversity_penalty
+                    )
+                )
+            if self.length_penalty != 1.0:
+                raise ValueError(
+                    single_beam_wrong_parameter_msg.format(flag_name="length_penalty", flag_value=self.length_penalty)
+                )
+            if self.constraints is not None:
+                raise ValueError(
+                    single_beam_wrong_parameter_msg.format(flag_name="constraints", flag_value=self.constraints)
+                )
+
+        # 3. detect incorrect paramaterization specific to advanced beam modes
+        else:
+            # constrained beam search
+            if self.constraints is not None:
+                constrained_wrong_parameter_msg = (
+                    "`constraints` is not `None`, triggering constrained beam search. However, {flag_name} is set to "
+                    "{flag_value}, which is incompatible with this generation mode. Set `constraints=None` or unset "
+                    "{flag_name} to continue."
+                )
+                if self.do_sample is True:
+                    raise ValueError(
+                        constrained_wrong_parameter_msg.format(flag_name="do_sample", flag_value=self.do_sample)
+                    )
+                if self.num_beam_groups != 1:
+                    raise ValueError(
+                        constrained_wrong_parameter_msg.format(
+                            flag_name="num_beam_groups", flag_value=self.num_beam_groups
+                        )
+                    )
+            # group beam search
+            if self.diversity_penalty != 0.0 or self.num_beam_groups != 1:
+                group_error_prefix = (
+                    "`diversity_penalty` is not 0.0 or `num_beam_groups` is not 1, triggering group beam search. In "
+                    "this generation mode, "
+                )
+                if self.do_sample is True:
+                    raise ValueError(group_error_prefix + "`do_sample` must be set to `False`")
+                if self.num_beams % self.num_beam_groups != 0:
+                    raise ValueError(group_error_prefix + "`num_beams` should be divisible by `num_beam_groups`")
+                if self.diversity_penalty == 0.0:
+                    raise ValueError(
+                        group_error_prefix
+                        + "`diversity_penalty` should be greater than `0.0`, otherwise your groups will be identical."
+                    )
+
+        # 4. check `num_return_sequences`
+        if self.num_return_sequences != 1:
+            if self.num_beams == 1:
+                if self.do_sample is False:
+                    raise ValueError(
+                        "Greedy methods without beam search do not support `num_return_sequences` different than 1 "
+                        f"(got {self.num_return_sequences})."
+                    )
+            elif self.num_return_sequences > self.num_beams:
+                raise ValueError(
+                    f"`num_return_sequences` ({self.num_return_sequences}) has to be smaller or equal to `num_beams` "
+                    f"({self.num_beams})."
+                )
+
     def save_pretrained(
         self,
         save_directory: Union[str, os.PathLike],

src/transformers/generation/utils.py

@@ -1493,13 +1493,6 @@ class GenerationMixin:
         # 7. determine generation mode
         generation_mode = self._get_generation_mode(generation_config, assistant_model)
 
-        if generation_config.num_beam_groups > generation_config.num_beams:
-            raise ValueError("`num_beam_groups` has to be smaller or equal to `num_beams`")
-        if generation_mode == GenerationMode.GROUP_BEAM_SEARCH and generation_config.do_sample is True:
-            raise ValueError(
-                "Diverse beam search cannot be used in sampling mode. Make sure that `do_sample` is set to `False`."
-            )
-
         if streamer is not None and (generation_config.num_beams > 1):
             raise ValueError(
                 "`streamer` cannot be used with beam search (yet!). Make sure that `num_beams` is set to 1."
@@ -1572,12 +1565,6 @@ class GenerationMixin:
                 **model_kwargs,
             )
         if generation_mode == GenerationMode.GREEDY_SEARCH:
-            if generation_config.num_return_sequences > 1:
-                raise ValueError(
-                    "num_return_sequences has to be 1 when doing greedy search, "
-                    f"but is {generation_config.num_return_sequences}."
-                )
-
             # 11. run greedy search
             return self.greedy_search(
                 input_ids,
@@ -1593,11 +1580,6 @@ class GenerationMixin:
             )
 
         elif generation_mode == GenerationMode.CONTRASTIVE_SEARCH:
-            if generation_config.num_return_sequences > 1:
-                raise ValueError(
-                    "num_return_sequences has to be 1 when doing contrastive search, "
-                    f"but is {generation_config.num_return_sequences}."
-                )
             if not model_kwargs["use_cache"]:
                 raise ValueError("Contrastive search requires `use_cache=True`")
 
@@ -1645,12 +1627,6 @@ class GenerationMixin:
             )
 
         elif generation_mode == GenerationMode.BEAM_SEARCH:
-            if generation_config.num_return_sequences > generation_config.num_beams:
-                raise ValueError("`num_return_sequences` has to be smaller or equal to `num_beams`.")
-
-            if stopping_criteria.max_length is None:
-                raise ValueError("`max_length` needs to be a stopping_criteria for now.")
-
             # 11. prepare beam search scorer
             beam_scorer = BeamSearchScorer(
                 batch_size=batch_size,
@@ -1686,8 +1662,6 @@ class GenerationMixin:
             # 11. prepare logits warper
             logits_warper = self._get_logits_warper(generation_config)
 
-            if stopping_criteria.max_length is None:
-                raise ValueError("`max_length` needs to be a stopping_criteria for now.")
             # 12. prepare beam search scorer
             beam_scorer = BeamSearchScorer(
                 batch_size=batch_size * generation_config.num_return_sequences,
@@ -1722,24 +1696,6 @@ class GenerationMixin:
             )
 
         elif generation_mode == GenerationMode.GROUP_BEAM_SEARCH:
-            if generation_config.num_return_sequences > generation_config.num_beams:
-                raise ValueError("`num_return_sequences` has to be smaller or equal to `num_beams`.")
-
-            if generation_config.num_beams % generation_config.num_beam_groups != 0:
-                raise ValueError("`num_beams` should be divisible by `num_beam_groups` for group beam search.")
-
-            if generation_config.diversity_penalty == 0.0:
-                raise ValueError(
-                    "`diversity_penalty` should be greater than `0.0`, otherwise your beam groups will be identical."
-                )
-
-            if stopping_criteria.max_length is None:
-                raise ValueError("`max_length` needs to be a stopping_criteria for now.")
-
-            has_default_typical_p = kwargs.get("typical_p") is None and generation_config.typical_p == 1.0
-            if not has_default_typical_p:
-                raise ValueError("Decoder argument `typical_p` is not supported with beam groups.")
-
             # 11. prepare beam search scorer
             beam_scorer = BeamSearchScorer(
                 batch_size=batch_size,
@@ -1773,21 +1729,6 @@ class GenerationMixin:
             )
 
         elif generation_mode == GenerationMode.CONSTRAINED_BEAM_SEARCH:
-            if generation_config.num_return_sequences > generation_config.num_beams:
-                raise ValueError("`num_return_sequences` has to be smaller or equal to `num_beams`.")
-
-            if stopping_criteria.max_length is None:
-                raise ValueError("`max_length` needs to be a stopping_criteria for now.")
-
-            if generation_config.num_beams <= 1:
-                raise ValueError("`num_beams` needs to be greater than 1 for constrained generation.")
-
-            if generation_config.do_sample:
-                raise ValueError("`do_sample` needs to be false for constrained generation.")
-
-            if generation_config.num_beam_groups is not None and generation_config.num_beam_groups > 1:
-                raise ValueError("`num_beam_groups` not supported yet for constrained generation.")
-
             final_constraints = []
             if generation_config.constraints is not None:
                 final_constraints = generation_config.constraints