story_cloze

  • বর্ণনা :

স্টোরি ক্লোজ টেস্ট হল গল্প বোঝা, গল্প তৈরি এবং স্ক্রিপ্ট শেখার মূল্যায়নের জন্য একটি নতুন কমনসেন্স যুক্তি কাঠামো। এই পরীক্ষার জন্য একটি চার-বাক্যের গল্পের সঠিক সমাপ্তি বেছে নেওয়ার জন্য একটি সিস্টেম প্রয়োজন।

  • অতিরিক্ত ডকুমেন্টেশন : কোড সহ কাগজপত্রে অন্বেষণ করুন

  • কনফিগারেশনের বিবরণ : 2018 সাল

  • হোমপেজ : https://www.cs.rochester.edu/nlp/rocstories/

  • সোর্স কোড : tfds.datasets.story_cloze.Builder

  • সংস্করণ :

    • 1.0.0 (ডিফল্ট): প্রাথমিক প্রকাশ।
  • ডাউনলোড আকার : Unknown size

  • ম্যানুয়াল ডাউনলোডের নির্দেশাবলী : এই ডেটাসেটের জন্য আপনাকে ডাউনলোড_config.manual_dir-এ ম্যানুয়ালি উৎস ডেটা download_config.manual_dir করতে হবে ( ~/tensorflow_datasets/downloads/manual/ ডিফল্ট):
    https://www.cs.rochester.edu/nlp/rocstories/ এ যান এবং ডেটাসেটগুলি পেতে গুগল ফর্মটি পূরণ করুন৷ আপনি ডেটাসেটগুলি ডাউনলোড করার লিঙ্ক সহ একটি ইমেল পাবেন। 2016 ডেটার জন্য, বৈধতা এবং পরীক্ষা ফাইলের নাম পরিবর্তন করে যথাক্রমে cloze_test val _spring2016.csv এবং cloze_test test _spring2016.csv রাখতে হবে। 2018 সংস্করণের জন্য, বৈধতা এবং পরীক্ষা ফাইলের নাম পরিবর্তন করে যথাক্রমে cloze_test val _winter2018.csv এবং cloze_test test _winter2018.csv রাখা দরকার। এই দুটি ফাইলকে ম্যানুয়াল ডিরেক্টরিতে সরান।

  • স্বয়ংক্রিয় ক্যাশে ( ডকুমেন্টেশন ): হ্যাঁ

  • বৈশিষ্ট্য গঠন :

FeaturesDict({
    'context': Text(shape=(), dtype=string),
    'endings': Sequence(Text(shape=(), dtype=string)),
    'label': int32,
})
  • বৈশিষ্ট্য ডকুমেন্টেশন :
বৈশিষ্ট্য ক্লাস আকৃতি ডিটাইপ বর্ণনা
ফিচারসডিক্ট
প্রসঙ্গ পাঠ্য স্ট্রিং
শেষ ক্রম (পাঠ্য) (কোনটিই নয়,) স্ট্রিং
লেবেল টেনসর int32
  • তত্ত্বাবধান করা কী (দেখুন as_supervised doc ): None

  • চিত্র ( tfds.show_examples ): সমর্থিত নয়।

  • উদ্ধৃতি :

@inproceedings{sharma-etal-2018-tackling,
    title = "Tackling the Story Ending Biases in The Story Cloze Test",
    author = "Sharma, Rishi  and
      Allen, James  and
      Bakhshandeh, Omid  and
      Mostafazadeh, Nasrin",
    booktitle = "Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)",
    month = jul,
    year = "2018",
    address = "Melbourne, Australia",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/P18-2119",
    doi = "10.18653/v1/P18-2119",
    pages = "752--757",
    abstract = "The Story Cloze Test (SCT) is a recent framework for evaluating story comprehension and script learning. There have been a variety of models tackling the SCT so far. Although the original goal behind the SCT was to require systems to perform deep language understanding and commonsense reasoning for successful narrative understanding, some recent models could perform significantly better than the initial baselines by leveraging human-authorship biases discovered in the SCT dataset. In order to shed some light on this issue, we have performed various data analysis and analyzed a variety of top performing models presented for this task. Given the statistics we have aggregated, we have designed a new crowdsourcing scheme that creates a new SCT dataset, which overcomes some of the biases. We benchmark a few models on the new dataset and show that the top-performing model on the original SCT dataset fails to keep up its performance. Our findings further signify the importance of benchmarking NLP systems on various evolving test sets.",
}

story_cloze/2016 (ডিফল্ট কনফিগারেশন)

  • ডেটাসেটের আকার : 1.15 MiB

  • বিভাজন :

বিভক্ত উদাহরণ
'test' 1,871
'validation' 1,871

story_cloze/2018

  • ডেটাসেটের আকার : 1015.04 KiB

  • বিভাজন :

বিভক্ত উদাহরণ
'test' 1,571
'validation' 1,571