Fix Trainer._symbolic_build (keras-team#19945)

keras/src/trainers/trainer.py

@@ -1001,12 +1001,6 @@ def _symbolic_build(self, iterator=None, data_batch=None):
  optimizer_unbuilt = (
  self.optimizer is not None and not self.optimizer.built
  )
- if model_unbuilt or compile_metrics_unbuilt or optimizer_unbuilt:
- if data_batch is None:
- for _, data in iterator.enumerate_epoch():
- data_batch = data[0]
- break
-
  if model_unbuilt or compile_metrics_unbuilt:
  # Create symbolic tensors matching an input batch.
 
@@ -1017,6 +1011,10 @@ def to_symbolic_input(v):
  v.shape, backend.standardize_dtype(v.dtype)
  )
 
+ if data_batch is None:
+ for _, data in iterator.enumerate_epoch():
+ data_batch = data[0]
+ break
  data_batch = tree.map_structure(to_symbolic_input, data_batch)
  (
  x,