Change tensor.data to tensor.detach() due to
pytorch/pytorch#6990 (comment)
tensor.detach() is more robust than tensor.data.

When run
bazel run --define MEDIAPIPE_DISABLE_GPU=1 mediapipe/examples/desktop/hello_world:hello_world
Downloading org_tensorflow will time out.
I want use the pre-downloaded org_tensorflow by putting the tar.gz file in the .cache corresponding directory, but it didn't work.How should I operate it correctly?

Current pytorch implementation ignores the argument split_f in the function train_batch_ch13 as shown below.

def train_batch_ch13(net, X, y, loss, trainer, devices):
    if isinstance(X, list):
        # Required for BERT Fine-tuning (to be covered later)
        X = [x.to(devices[0]) for x in X]
    else:
        X = X.to(devices[0])
...

Todo: Define the argument `