:param config: :param is_training: :param input_ids: Tensor thats of size [batch_size, seq_length] :param cache: Optionally, a tensor to use that will contain cached information of the size [batch_size, num_layers, 2, num_heads, cache_length, features]
(self,
config: GroverConfig,
is_training,
input_ids,
cache=None,
do_cache=False,
pad_token_id=0,
chop_off_last_token=True,
scope=None,
reuse=False)
source not stored for this graph (policy: none)
nothing calls this directly
no test coverage detected