当前位置: 首页>>代码示例>>Python>>正文


Python BigML.download_dataset方法代码示例

本文整理汇总了Python中bigml.api.BigML.download_dataset方法的典型用法代码示例。如果您正苦于以下问题:Python BigML.download_dataset方法的具体用法?Python BigML.download_dataset怎么用?Python BigML.download_dataset使用的例子?那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在bigml.api.BigML的用法示例。


在下文中一共展示了BigML.download_dataset方法的1个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Python代码示例。

示例1: Cluster

# 需要导入模块: from bigml.api import BigML [as 别名]
# 或者: from bigml.api.BigML import download_dataset [as 别名]

#.........这里部分代码省略.........
        for point in list_of_points:
            centroid_id = None
            if isinstance(point, Centroid):
                centroid_id = point.centroid_id
                point = point.center
                by_name = False
            clean_point, unique_terms = self._prepare_for_distance( \
                point, by_name=by_name)
            if clean_point != reference_point:
                result = {"data": point, "distance": reference.distance2( \
                    clean_point, unique_terms, self.scales)}
                if centroid_id is not None:
                    result.update({"centroid_id": centroid_id})
                distances.append(result)
        return distances

    def points_in_cluster(self, centroid_id):
        """Returns the list of data points that fall in one cluster.

        """

        cluster_datasets = self.datasets
        centroid_dataset = cluster_datasets.get(centroid_id)
        if self.api is None:
            self.api = BigML(storage=STORAGE)
        if centroid_dataset in [None, ""]:
            centroid_dataset = self.api.create_dataset( \
                self.resource_id, {"centroid": centroid_id})
            self.api.ok(centroid_dataset)
        else:
            centroid_dataset = self.api.check_resource( \
                "dataset/%s" % centroid_dataset)
        # download dataset to compute local predictions
        downloaded_data = self.api.download_dataset( \
            centroid_dataset["resource"])
        if PY3:
            text_reader = codecs.getreader("utf-8")
            downloaded_data = text_reader(downloaded_data)
        reader = csv.DictReader(downloaded_data)
        points = []
        for row in reader:
            points.append(row)
        return points

    def closest_in_cluster(self, reference_point,
                           number_of_points=None,
                           centroid_id=None,
                           by_name=True):
        """Computes the list of data points closer to a reference point.
        If no centroid_id information is provided, the points are chosen
        from the same cluster as the reference point.
        The points are returned in a list, sorted according
        to their distance to the reference point. The number_of_points
        parameter can be set to truncate the list to a maximum number of
        results. The response is a dictionary that contains the
        centroid id of the cluster plus the list of points
        """
        if centroid_id is not None and centroid_id not in \
                [centroid.centroid_id for centroid in self.centroids]:
            raise AttributeError( \
                "Failed to find the provided centroid_id: %s" % centroid_id)
        if centroid_id is None:
            # finding the reference point cluster's centroid
            centroid_info = self.centroid(reference_point, by_name=True)
            centroid_id = centroid_info["centroid_id"]
        # reading the points that fall in the same cluster
开发者ID:charleslparker,项目名称:python,代码行数:70,代码来源:cluster.py


注:本文中的bigml.api.BigML.download_dataset方法示例由纯净天空整理自Github/MSDocs等开源代码及文档管理平台,相关代码片段筛选自各路编程大神贡献的开源项目,源码版权归原作者所有,传播和使用请参考对应项目的License;未经允许,请勿转载。