ENH [`AutoQuantizer`]: enhance trainer + not supported quant methods (#28991)

* enhance trainer + not support quant methods

* remove all old logic

* add version

src/transformers/modeling_utils.py

@@ -4190,6 +4190,18 @@ class PreTrainedModel(nn.Module, ModuleUtilsMixin, GenerationMixin, PushToHubMix
 
             logger.warning_once(warn_string)
 
+    @property
+    def _is_quantized_training_enabled(self):
+        logger.warning(
+            "`_is_quantized_training_enabled` is going to be deprecated in transformers 4.39.0. Please use `model.hf_quantizer.is_trainable` instead",
+            FutureWarning,
+        )
+
+        if not hasattr(self, "hf_quantizer"):
+            return False
+
+        return self.hf_quantizer.is_trainable
+
 
 PreTrainedModel.push_to_hub = copy_func(PreTrainedModel.push_to_hub)
 if PreTrainedModel.push_to_hub.__doc__ is not None:

src/transformers/quantizers/base.py

@@ -176,7 +176,6 @@ class HfQuantizer(ABC):
             kwargs (`dict`, *optional*):
                 The keyword arguments that are passed along `_process_model_after_weight_loading`.
         """
-        model._is_quantized_training_enabled = self.is_trainable
         return self._process_model_after_weight_loading(model, **kwargs)
 
     @abstractmethod

src/transformers/quantizers/quantizer_bnb_4bit.py

@@ -289,7 +289,6 @@ class Bnb4BitHfQuantizer(HfQuantizer):
 
     # Copied from transformers.quantizers.quantizer_bnb_8bit.Bnb8BitHfQuantizer._process_model_after_weight_loading with 8bit->4bit
     def _process_model_after_weight_loading(self, model: "PreTrainedModel", **kwargs):
-        model._is_quantized_training_enabled = self.is_trainable
         model.is_loaded_in_4bit = True
         model.is_4bit_serializable = self.is_serializable
         return model

src/transformers/quantizers/quantizer_bnb_8bit.py

@@ -205,7 +205,6 @@ class Bnb8BitHfQuantizer(HfQuantizer):
             unexpected_keys.remove(fp16_statistics_key)
 
     def _process_model_after_weight_loading(self, model: "PreTrainedModel", **kwargs):
-        model._is_quantized_training_enabled = self.is_trainable
         model.is_loaded_in_8bit = True
         model.is_8bit_serializable = self.is_serializable
         return model

src/transformers/trainer.py

@@ -420,6 +420,9 @@ class Trainer:
         _is_quantized_and_base_model = getattr(model, "is_quantized", False) and not getattr(
             model, "_hf_peft_config_loaded", False
         )
+        _quantization_method_supports_training = (
+            getattr(model, "hf_quantizer", None) is not None and model.hf_quantizer.is_trainable
+        )
 
         # At this stage the model is already loaded
         if _is_quantized_and_base_model and not _is_peft_model(model):
@@ -428,10 +431,11 @@ class Trainer:
                 " the quantized model to correctly perform fine-tuning. Please see: https://huggingface.co/docs/transformers/peft"
                 " for more details"
             )
-        elif _is_quantized_and_base_model and not getattr(model, "_is_quantized_training_enabled", False):
+        elif _is_quantized_and_base_model and not _quantization_method_supports_training:
             raise ValueError(
-                "The model you want to train is loaded in 8-bit precision.  if you want to fine-tune an 8-bit"
-                " model, please make sure that you have installed `bitsandbytes>=0.37.0`. "
+                f"The model you are trying to fine-tune is quantized with {model.hf_quantizer.quantization_config.quant_method}"
+                " but that quantization method do not support training. Please open an issue on GitHub: https://github.com/huggingface/transformers"
+                f" to request the support for training support for {model.hf_quantizer.quantization_config.quant_method}"
             )
 
         self.is_fsdp_xla_enabled = args.fsdp_config["xla"]