Chronos-2: Add option to remove PrinterCallback (#410)

*Issue #, if available:*

*Description of changes:*


By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice.

src/chronos/chronos2/pipeline.py

@@ -19,7 +19,6 @@ from transformers import AutoConfig
 from transformers.utils.import_utils import is_peft_available
 from transformers.utils.peft_utils import find_adapter_config_file
 
-
 import chronos.chronos2
 from chronos.base import BaseChronosPipeline, ForecastType
 from chronos.chronos2 import Chronos2Model
@@ -114,6 +113,7 @@ class Chronos2Pipeline(BaseChronosPipeline):
         min_past: int | None = None,
         finetuned_ckpt_name: str = "finetuned-ckpt",
         callbacks: list["TrainerCallback"] | None = None,
+        remove_printer_callback: bool = False,
         **extra_trainer_kwargs,
     ) -> "Chronos2Pipeline":
         """
@@ -156,6 +156,8 @@ class Chronos2Pipeline(BaseChronosPipeline):
             The name of the directory inside `output_dir` in which the final fine-tuned checkpoint will be saved, by default "finetuned-ckpt"
         callbacks
             A list of `TrainerCallback`s which will be forwarded to the HuggingFace `Trainer`
+        remove_printer_callback
+            If True, all instances of `PrinterCallback` are removed from callbacks
         **extra_trainer_kwargs
             Extra kwargs are directly forwarded to `TrainingArguments`
 
@@ -165,6 +167,7 @@ class Chronos2Pipeline(BaseChronosPipeline):
         """
 
         import torch.cuda
+        from transformers.trainer_callback import PrinterCallback
         from transformers.training_args import TrainingArguments
 
         if finetune_mode == "lora":
@@ -322,6 +325,10 @@ class Chronos2Pipeline(BaseChronosPipeline):
             eval_dataset=eval_dataset,
             callbacks=callbacks,
         )
+
+        if remove_printer_callback:
+            trainer.pop_callback(PrinterCallback)
+
         trainer.train()
 
         # update max_output_patches, if the model was fine-tuned with longer prediction_length