[Cherry Pick] allow dataset size smaller than calibration samples (#2091

) (#2179)

* allow dataset size smaller than calibration samples (#2091)

* merge issue

src/sparseml/transformers/finetune/data/data_helpers.py

@@ -49,9 +49,17 @@ def format_calibration_data(
     :param accelerator: optional accelerator for if preparing in FSDP mode
     :return: list of trimmed calibration data tensors
     """
-    num_calibration_samples = num_calibration_samples or len(tokenized_dataset)
+    safe_calibration_samples = len(tokenized_dataset)
+    if num_calibration_samples is not None:
+        safe_calibration_samples = min(len(tokenized_dataset), num_calibration_samples)
+        if safe_calibration_samples != num_calibration_samples:
+            LOGGER.warn(
+                f"Requested {num_calibration_samples} calibration samples but "
+                f"the provided dataset only has {safe_calibration_samples}. "
+            )
+
     shuffled_calibration = tokenized_dataset.shuffle()
-    shuffled_calibration = shuffled_calibration.select(range(num_calibration_samples))
+    shuffled_calibration = shuffled_calibration.select(range(safe_calibration_samples))
 
     dataloader_params = {
         "batch_size": 1,