zhenghong issues

Repositories
Issues
Comments

Results 1 issues of


                                            zhenghong

tf 1.15 use tensorflow-io read oss file, then return empty line data

in code： ```python …. _NUM_PARALLEL_CALLS = 8 dataset = dataset.batch(self._data_config.batch_size) dataset = dataset.map(self._parse_csv, num_parallel_calls=_NUM_PARALLEL_CALLS) dataset = dataset.map(map_func=self._preprocess, num_parallel_calls=_NUM_PARALLEL_CALLS) dataset = dataset.prefetch(buffer_size=self._prefetch_size) … def _parse_csv(self, line): record_defaults = [ self.get_type_defaults(t, v)...