How to change data transform on built-in dataset?

I've read how you implement cifar10 in
- `fedml/data/data_loader.py` : Entry point of loading dataset
- `fedml/data/cifar10/efficient_loader.py` : loading cifar10 efficiently

I'm wondering if it is possible to change `_data_transforms` to a customized one? Can I configure it in `fedml_config.yaml`?

Since I want to perform federated transfer learning (FTL) from the ImageNet-pretrained model,
which have been trained at different resolution (224x224).

Thanks!

```python
def _data_transforms_cifar10():
CIFAR_MEAN = [0.49139968, 0.48215827, 0.44653124]
CIFAR_STD = [0.24703233, 0.24348505, 0.26158768]

train_transform = transforms.Compose(
[
transforms.ToPILImage(),
transforms.RandomCrop(32, padding=4),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize(CIFAR_MEAN, CIFAR_STD),
]
)

train_transform.transforms.append(Cutout(16))

valid_transform = transforms.Compose(
[
transforms.ToTensor(),
transforms.Normalize(CIFAR_MEAN, CIFAR_STD),
]
)

return train_transform, valid_transform

def load_cifar10_data(datadir):
train_transform, test_transform = _data_transforms_cifar10()

cifar10_train_ds = CIFAR10_truncated(
datadir, train=True, download=True, transform=train_transform
)
cifar10_test_ds = CIFAR10_truncated(
datadir, train=False, download=True, transform=test_transform
)

X_train, y_train = cifar10_train_ds.data, cifar10_train_ds.target
X_test, y_test = cifar10_test_ds.data, cifar10_test_ds.target

return (X_train, y_train, X_test, y_test)
```