---
language:
- en

---
# AutoTrain Dataset for project: legal-data

## Dataset Description

This dataset has been automatically processed by AutoTrain for project legal-data.

### Languages

The BCP-47 code for the dataset's language is en.

## Dataset Structure

### Data Instances

A sample from this dataset looks as follows:

```json
[
  {
    "context": "Merely opening the door to a police\nofficer does not constitute consent to entry and\nsearch. Thus, whatever such a search turns\nup would be inadmissible in evidence. Of\ncourse, if contraband or evidence of a crime is\nin \u201cplain view\u201d from the doorway, the officer\nmay seize it.",
    "question": "If I agree to open my door to talk\nto a police officer, and the officer\nenters without my permission\nand searches, is the search valid?",
    "answers.text": [
      "No. Merely opening the door to a police\nofficer does not constitute consent to entry and\nsearch. Thus, whatever such a search turns\nup would be inadmissible in evidence. Of\ncourse, if contraband or evidence of a crime is\nin \u201cplain view\u201d from the doorway, the officer\nmay seize it. (See Section IV, below.)"
    ],
    "answers.answer_start": [
      0
    ]
  },
  {
    "context": "Most people\ninstinctively understand the concept of\nprivacy. It is the freedom to decide which\ndetails of your life shall be revealed to the\npublic and which shall be revealed only\nto those you care to share them with. To\nhonor this freedom, the Fourth Amendment\nprotects against \u201cunreasonable\u201d searches\nand seizures by state or federal law\nenforcement authorities. However, the\nFourth Amendment does not protect against\nsearches initiated by nongovernmental\npeople, such as employers, landlords, and\nprivate security personnel, unless the search\nis made at the behest of a law enforcement\nauthority.",
    "question": "What are the search and seizure\nprovisions of the Fourth Amendment\nall about?",
    "answers.text": [
      "They are about privacy."
    ],
    "answers.answer_start": [
      0
    ]
  }
]
```

### Dataset Fields

The dataset has the following fields (also called "features"):

```json
{
  "context": "Value(dtype='string', id=None)",
  "question": "Value(dtype='string', id=None)",
  "answers.text": "Sequence(feature=Value(dtype='string', id=None), length=-1, id=None)",
  "answers.answer_start": "Sequence(feature=Value(dtype='int32', id=None), length=-1, id=None)"
}
```

### Dataset Splits

This dataset is split into a train and validation split. The split sizes are as follow:

| Split name   | Num samples         |
| ------------ | ------------------- |
| train        | 40 |
| valid        | 10 |