{
    "info": {
        "author": "Taycir Yahmed",
        "author_email": "taycir.yahmed@gmail.com",
        "bugtrack_url": null,
        "classifiers": [
            "Operating System :: OS Independent",
            "Programming Language :: Python :: 3"
        ],
        "description": "# mcQA : Multiple Choice Questions Answering \n\nAnswering multiple choice questions with Language Models.\n\n[![CircleCI](https://circleci.com/gh/mcQA-suite/mcQA.svg?style=svg)](https://circleci.com/gh/mcQA-suite/mcQA)\n[![PyPI Version](https://img.shields.io/pypi/v/mcqa.svg)](https://pypi.org/project/mcqa/)\n![GitHub](https://img.shields.io/github/license/mcqa-suite/mcqa.svg)\n[![codecov](https://codecov.io/gh/mcqa-suite/mcQA/branch/master/graph/badge.svg)](https://codecov.io/gh/mcqa-suite/mcQA)\n[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](http://makeapullrequest.com)\n\n## Installation\n\n### With pip\n\n```shell\npip install mcqa\n```\n\n### From source\n\n```shell\ngit clone https://github.com/mcqa-suite/mcqa.git\ncd mcQA\npip install -e .\n```\n\n## Getting started\n\n### Data preparation\n\nTo train a `mcQA` model, you need to create a csv file with n+2 columns, n being the number of choices for each question. The first column should be the context sentence, the n following columns should be the choices for that question and the last column is the selected answer. \n\nBelow is an example of a 3 choice question (taken from the [CoS-E dataset](https://arxiv.org/pdf/1906.02361.pdf)) :\n\n| Context sentence  | Choice 1                | Choice 2            | Choice 3    | Label|\n| ----------------- | --------------------|--------------------|--------------------|-------------|\n| People do what during their time off from work?| take trips | brow shorter | become hysterical | take trips |\n\nIf you have a trained `mcQA` model and want to infer on a dataset, it should have the same format as the train data, but the `label` column. \n\nSee example data preparation below:\n\n```python\nfrom mcqa.data import MCQAData\n\nmcqa_data = MCQAData(bert_model=\"bert-base-uncased\", \n                     lower_case=True, \n                     max_seq_length=256) \n                     \ntrain_dataset = mcqa_data.read(data_file='swagaf/data/train.csv', is_training=True)\ntest_dataset = mcqa_data.read(data_file='swagaf/data/test.csv', is_training=False)\n```\n\n### Model training \n\n```python\nfrom mcqa.models import Model\n\nmdl = Model(bert_model=\"bert-base-uncased\",\n            device=\"cuda\") \n            \nmdl.fit(train_dataset, \n        train_batch_size=32, \n        num_train_epochs=20)\n```\n\n### Prediction\n\n```python\npreds = mdl.predict(test_dataset, \n                    eval_batch_size=32)\n```\n\n### Evaluation\n\n```python\nfrom sklearn.metrics import accuracy_score\nfrom mcqa.data import get_labels\n\nprint(accuracy_score(preds, get_labels(train_dataset)))\n```\n## References\n\n| Type                 | Title                                                                                                                                        | Author                                                                                 | Year |\n| -------------------- | -------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------- | ---- |\n|:newspaper: Paper| [Explain Yourself! Leveraging Language Models for Commonsense Reasoning](https://arxiv.org/pdf/1906.02361.pdf)|Nazneen Fatema Rajani, Bryan McCann, Caiming Xiong and Richard Socher| ACL 2019|\n|:newspaper: Paper|[SWAG: A Large-Scale Adversarial Dataset for Grounded Commonsense Inference](https://arxiv.org/abs/1808.05326)|Rowan Zellers, Yonatan Bisk, Roy Schwartz and Yejin Choi|2018|",
        "description_content_type": "text/markdown",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/mcQA-suite/mcQA",
        "keywords": "",
        "license": "",
        "maintainer": "",
        "maintainer_email": "",
        "name": "mcqa",
        "package_url": "https://pypi.org/project/mcqa/",
        "platform": "",
        "project_url": "https://pypi.org/project/mcqa/",
        "project_urls": {
            "Homepage": "https://github.com/mcQA-suite/mcQA"
        },
        "release_url": "https://pypi.org/project/mcqa/0.1.1/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "Answering multiple choice questions with Language Models.",
        "version": "0.1.1"
    },
    "last_serial": 5687355,
    "releases": {
        "0.0.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "f353580fbccce6655865abb773dcc773",
                    "sha256": "0200cdae704c3d77641ddc416f53661e2c080f99470a11bc9802d682178c2689"
                },
                "downloads": -1,
                "filename": "mcqa-0.0.1-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "f353580fbccce6655865abb773dcc773",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 5490,
                "upload_time": "2019-07-15T20:53:03",
                "url": "https://files.pythonhosted.org/packages/9c/7d/0d66a36559901430d876b5c9151d44b4ef39fc4f25a870edc30971656b5b/mcqa-0.0.1-py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "dfaae4bf1c4152ecb4c8251fb7017b3c",
                    "sha256": "6dffbf103356cf77b8f08254f7494b15bc09ebf20fbbf1d5584498f2944b8207"
                },
                "downloads": -1,
                "filename": "mcqa-0.0.1.tar.gz",
                "has_sig": false,
                "md5_digest": "dfaae4bf1c4152ecb4c8251fb7017b3c",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 1143,
                "upload_time": "2019-07-15T20:53:05",
                "url": "https://files.pythonhosted.org/packages/da/c2/95b6db4fdddb5a2a026f9876bb52b8dd80b6765f11fae399b18502e07ad0/mcqa-0.0.1.tar.gz"
            }
        ],
        "0.1.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "eeb2b4a12b625fdb83e4ddf96f4815de",
                    "sha256": "11d4966c82558aed83d6be9531c0d7e8c2b90c2f58f48bbb50f7e0a79460276f"
                },
                "downloads": -1,
                "filename": "mcqa-0.1.0.tar.gz",
                "has_sig": false,
                "md5_digest": "eeb2b4a12b625fdb83e4ddf96f4815de",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 13756,
                "upload_time": "2019-07-26T09:22:57",
                "url": "https://files.pythonhosted.org/packages/b5/d7/1fa25c55e7cd0b4cc101ce4ddac2b811fe579253573c42f274b78cdd7b59/mcqa-0.1.0.tar.gz"
            }
        ],
        "0.1.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "2664ce9f09c18ae93148eaa7fa582fa2",
                    "sha256": "7317fcf33d232cf42f70a097d9847e4f77433329188bff7635491387dcf5af13"
                },
                "downloads": -1,
                "filename": "mcqa-0.1.1.tar.gz",
                "has_sig": false,
                "md5_digest": "2664ce9f09c18ae93148eaa7fa582fa2",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 15621,
                "upload_time": "2019-08-16T12:22:26",
                "url": "https://files.pythonhosted.org/packages/54/a7/92e02e08b640c6b4f339b249211688fc52b09ef32efb603527b766e4f0f2/mcqa-0.1.1.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "2664ce9f09c18ae93148eaa7fa582fa2",
                "sha256": "7317fcf33d232cf42f70a097d9847e4f77433329188bff7635491387dcf5af13"
            },
            "downloads": -1,
            "filename": "mcqa-0.1.1.tar.gz",
            "has_sig": false,
            "md5_digest": "2664ce9f09c18ae93148eaa7fa582fa2",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 15621,
            "upload_time": "2019-08-16T12:22:26",
            "url": "https://files.pythonhosted.org/packages/54/a7/92e02e08b640c6b4f339b249211688fc52b09ef32efb603527b766e4f0f2/mcqa-0.1.1.tar.gz"
        }
    ]
}