Displaying data from summarization dataset in TensorFlow (using TensorFlow datasets)

Question

I'm new to Machine Learning and a newbie when it comes to utilizing the TensorFlow Module in Python.

I'm currently working with summarization and the dataset library in TensorFlow has many convenient datasets available for training the summarizers. However, I wanted to take a look at their contents before chosing one in particular, does anyone know how to display the dataset as a Table in the Python console?

So far, I have the example code (for the Opinosis dataset) from the TensorFlow website, which is the following:

# Copyright 2022 The TensorFlow Datasets Authors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

"""Opinosis Opinion Dataset."""

import os

import tensorflow as tf
import tensorflow_datasets.public_api as tfds

_CITATION = """
@inproceedings{ganesan2010opinosis,
  title={Opinosis: a graph-based approach to abstractive summarization of highly redundant opinions},
  author={Ganesan, Kavita and Zhai, ChengXiang and Han, Jiawei},
  booktitle={Proceedings of the 23rd International Conference on Computational Linguistics},
  pages={340--348},
  year={2010},
  organization={Association for Computational Linguistics}
}
"""

_DESCRIPTION = """
The Opinosis Opinion Dataset consists of sentences extracted from reviews for 51 topics.
Topics and opinions are obtained from Tripadvisor, Edmunds.com and Amazon.com.
"""

_URL = "https://github.com/kavgan/opinosis-summarization/raw/master/OpinosisDataset1.0_0.zip"

_REVIEW_SENTS = "review_sents"
_SUMMARIES = "summaries"


class Opinosis(tfds.core.GeneratorBasedBuilder):
  """Opinosis Opinion Dataset."""

  VERSION = tfds.core.Version("1.0.0")

  def _info(self):
    return tfds.core.DatasetInfo(
        builder=self,
        description=_DESCRIPTION,
        features=tfds.features.FeaturesDict({
            _REVIEW_SENTS: tfds.features.Text(),
            _SUMMARIES: tfds.features.Sequence(tfds.features.Text())
        }),
        supervised_keys=(_REVIEW_SENTS, _SUMMARIES),
        homepage="http://kavita-ganesan.com/opinosis/",
        citation=_CITATION,
    )

  def _split_generators(self, dl_manager):
    """Returns SplitGenerators."""
    extract_path = dl_manager.download_and_extract(_URL)
    return [
        tfds.core.SplitGenerator(
            name=tfds.Split.TRAIN,
            gen_kwargs={"path": extract_path},
        ),
    ]

  def _generate_examples(self, path=None):
    """Yields examples."""
    topics_path = os.path.join(path, "topics")
    filenames = tf.io.gfile.listdir(topics_path)
    for filename in filenames:
      file_path = os.path.join(topics_path, filename)
      topic_name = filename.split(".txt")[0]
      with tf.io.gfile.GFile(file_path, "rb") as src_f:
        input_data = src_f.read()
      summaries_path = os.path.join(path, "summaries-gold", topic_name)
      summary_lst = []
      for summ_filename in sorted(tf.io.gfile.listdir(summaries_path)):
        file_path = os.path.join(summaries_path, summ_filename)
        with tf.io.gfile.GFile(file_path, "rb") as tgt_f:
          data = tgt_f.read().strip()
          summary_lst.append(data)
      summary_data = summary_lst
      yield filename, {_REVIEW_SENTS: input_data, _SUMMARIES: summary_data}```

Answer 1

That is the source code for the Opinosis dataset. You don't need to copy it over to your code. This should give you a good idea of how to use tensorflow datasets. Opinosis doesn't make much sense displayed as a table, so to get an idea of the contents I would just print a few examples. E.g:

import tensorflow_datasets as tfds
ds, info = tfds.load('opinosis', split='train', with_info=True)
ds_iter = iter(ds)
for i in range(3): 
  print(next(ds_iter))

If you really want to see a table, you can use:

print(tfds.as_dataframe(ds.take(3), info))