📄 En¶

CLAP NQ is created from the subset of Natural Questions (NQ) that have a long answer but no short answer. NQ consists of ~380k examples. There are ~30k questions that are long answers without short answers excluding tables and lists. To increases the likelihood of longer answers we only explored ones that have more than 5 sentences in the passage. The subset that was annotated consists of ~12k examples. All examples where cohesion of non-consecutive sentences was required for the answer were annotated a second time. The final dataset is made up of all data that went through two rounds of annotation. (We provide the single round annotations as well - it is only training data) An equal amount of unanswerable questions have also been added from the original NQ train/dev sets. Details about the annotation task and unanswerables can be found at https://github.com/primeqa/clapnq/blob/main/annotated_data.

Tags: license:Apache License 2.0, url:https://huggingface.co/datasets/PrimeQA/clapnq, category:dataset

cards.rag.benchmark.clap_nq.en

TaskCard(
    loader=LoadCSV(
        sep="   ",
        files={
            "train": "https://raw.githubusercontent.com/primeqa/clapnq/main/retrieval/train/question_train_answerable.tsv",
            "test": "https://raw.githubusercontent.com/primeqa/clapnq/main/retrieval/dev/question_dev_answerable.tsv",
        },
        data_classification_policy=[
            "public",
        ],
    ),
    preprocess_steps=[
        Copy(
            field_to_field={
                "question": "question",
                "id": "question_id",
            },
        ),
        ListFieldValues(
            fields=[
                "doc-id-list",
            ],
            to_field="reference_context_ids",
        ),
        ListFieldValues(
            fields=[
                "answers",
            ],
            to_field="reference_answers",
        ),
    ],
    task="tasks.rag.end_to_end",
    templates={
        "default": "templates.rag.end_to_end.json_predictions",
    },
)

[source]

from unitxt.loaders import LoadCSV
from unitxt.operators import Copy, ListFieldValues

Explanation about TaskCard¶

TaskCard delineates the phases in transforming the source dataset into model input, and specifies the metrics for evaluation of model output.

Args:

loader:
specifies the source address and the loading operator that can access that source and transform it into a unitxt multistream.

preprocess_steps:
list of unitxt operators to process the data source into model input.

task:
specifies the fields (of the already (pre)processed instance) making the inputs, the fields making the outputs, and the metrics to be used for evaluating the model output.

templates:
format strings to be applied on the input fields (specified by the task) and the output fields. The template also carries the instructions and the list of postprocessing steps, to be applied to the model output.

Explanation about Copy¶

Copies values from specified fields to specified fields.

Args (of parent class):
field_to_field (Union[List[List], Dict[str, str]]): A list of lists, where each sublist contains the source field and the destination field, or a dictionary mapping source fields to destination fields.

Examples:
An input instance {“a”: 2, “b”: 3}, when processed by Copy(field_to_field={"a": "b"}) would yield {“a”: 2, “b”: 2}, and when processed by Copy(field_to_field={"a": "c"}) would yield {“a”: 2, “b”: 3, “c”: 2}

with field names containing / , we can also copy inside the field: Copy(field="a/0",to_field="a") would process instance {“a”: [1, 3]} into {“a”: 1}

Explanation about ListFieldValues¶

Concatenates values of multiple fields into a list, and assigns it to a new field.

Explanation about LoadCSV¶

Loads data from CSV files.

Supports streaming and can handle large files by loading them in chunks.
Args:
files (Dict[str, str]): A dictionary mapping names to file paths. chunksize : Size of the chunks to load at a time. loader_limit: Optional integer to specify a limit on the number of records to load. streaming: Bool indicating if streaming should be used. sep: String specifying the separator used in the CSV files. indirect_read: Bool indicating if to open a remote file with urllib first column_names: Optional list of column names to use instead of header row.

Example:
Loading csv
load_csv = LoadCSV(files={'train': 'path/to/train.csv'}, chunksize=100)
Loading TSV with custom column names
load_csv = LoadCSV(
    files={'train': 'path/to/train.tsv'},
    sep='\t',
    column_names=['id', 'question', 'table_name', 'answer']
)

References: templates.rag.end_to_end.json_predictions, tasks.rag.end_to_end