Why do I need to add a dimension for data?

This code is from the Tensorflow website

def df_to_dataset(dataframe, shuffle=True, batch_size=32):
  df = dataframe.copy()
  labels = df.pop('target')
  df = {key: value[:,tf.newaxis] for key, value in dataframe.items()}
  ds = tf.data.Dataset.from_tensor_slices((dict(df), labels))
  if shuffle:
    ds = ds.shuffle(buffer_size=len(dataframe))
  ds = ds.batch(batch_size)
  ds = ds.prefetch(batch_size)
  return ds

My question is in this line:

df = {key: value[:,tf.newaxis] for key, value in dataframe.items()}

Why do I have to add a new axis (specifically value[:,tf.newaxis]) for every key value pair?