dense_vector/track.py [167:184]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
            self._iters = 0
        return result


# For each query this will generate both the knn query and the equivalent
# score script query. The two queries can then be executed and used
# to gauge the accuracy of the knn query.
class KnnRecallParamSource:
    def __init__(self, track, params, **kwargs):
        if len(track.indices) == 1:
            default_index = track.indices[0].name
        else:
            default_index = "_all"

        self._index_name = params.get("index", default_index)
        self._cache = params.get("cache", False)
        self._params = params
        self.infinite = True
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



msmarco-v2-vector/track.py [135:149]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
            self._iters = 0
        return result


class KnnRecallParamSource:
    def __init__(self, track, params, **kwargs):
        if len(track.indices) == 1:
            default_index = track.indices[0].name
        else:
            default_index = "_all"

        self._index_name = params.get("index", default_index)
        self._cache = params.get("cache", False)
        self._params = params
        self.infinite = True
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -