Repository navigation
Possible ByteVocabulary Bug #158
Description
464 def _decode_tf(self, ids):
465 """Decode in TensorFlow.
466
467 Args:
468 ids: a 1d tf.Tensor with dtype tf.int32
469 Returns:
470 a tf Scalar with dtype tf.string
471 """
472 return tf.py_function(func=self.decode, inp=[ids], Tout=tf.string)
The param 'ids' passed to _decode_tf above is a 1d tf.Tensor, and on line 472 it is wrapped into a list, but when self.decode is called with the param [ids], it throws the error in the list comprehension on line 100 (shown below):
File "/home/ubuntu/anaconda3/envs/google_t5/lib/python3.7/site-packages/seqio/vocabularies.py", line 100, in
decode
for i in clean_ids
File "/home/ubuntu/anaconda3/envs/google_t5/lib/python3.7/site-packages/seqio/vocabularies.py", line 100, in
for i in clean_ids
File "/home/ubuntu/anaconda3/envs/google_t5/lib/python3.7/site-packages/tensorflow/python/framework/ops.py",
line 1007, in bool
return bool(self._numpy())
ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()
92 def decode(self, ids: Iterable[int]):
93 """Detokenizes int32 iterable to a string, up through first EOS."""
94 clean_ids = list(ids)
95
96 if self.unk_id is not None:
97 vocab_size = self._base_vocab_size
98 clean_ids = [
99 self.unk_id if i >= vocab_size else i
100 for i in clean_ids
101 ]
102
103 if self.eos_id is not None and self.eos_id in clean_ids:
104 clean_ids = clean_ids[:clean_ids.index(self.eos_id) + 1]
105
106 return self._decode(clean_ids)
Activity
adarob commentedon Nov 8, 2021
@nconstant-google this does look like a bug
xwd commentedon Nov 8, 2021
This can be reproduced by running a pretrained byt5 model (from https://github.com/google-research/byt5) using the following command (I'm using GPU, I believe the error would persist with TPUs):
export BUCKET=~/data/t5/byt5-data/
export TASK=byt5_ner_multilingual
export MODEL_DIR="${BUCKET}/${TASK}"
python -m t5.models.mesh_transformer_main
--module_import="byt5.tasks"
--model_dir="${MODEL_DIR}"
--gin_file="${MODEL_DIR}/operative_config.gin"
--t5_tfds_data_dir="${BUCKET}/t5-tfds"
--gin_file="eval.gin"
--gin_file="beam_search.gin"
--gin_param="utils.run.mesh_shape = 'model:2,batch:1'"
--gin_param="utils.run.mesh_devices = ['gpu:0','gpu:1']"
--gin_param="run.dataset_split = 'validation'"
--gin_param="MIXTURE_NAME = 'byt5_ner_eval.ja'" \