glossary.test.json

{
    "source_id": "glossary",
    "url": "https://developers.google.com/machine-learning/glossary",
    "sections": [
        {
            "name": "class",
            "split": "test",
            "text": [
                [
                    "class",
                    "类别"
                ],
                [
                    "One of a set of enumerated target values for a label.",
                    "为标签枚举的一组目标值中的一个。"
                ],
                [
                    "For example, in a binary classification model that detects spam, the two classes are spam and not spam.",
                    "例如，在检测垃圾邮件的二元分类模型中，两种类别分别是“垃圾邮件”和“非垃圾邮件”。"
                ],
                [
                    "In a multi-class classification model that identifies dog breeds, the classes would be poodle, beagle, pug, and so on.",
                    "在识别狗品种的多类别分类模型中，类别可以是“贵宾犬”、“小猎犬”、“哈巴犬”等等。"
                ]
            ]
        },
        {
            "name": "classification model",
            "split": "test",
            "text": [
                [
                    "classification model",
                    "分类模型"
                ],
                [
                    "A type of machine learning model for distinguishing among two or more discrete classes.",
                    "一种机器学习模型，用于区分两种或多种离散类别。s"
                ],
                [
                    "For example, a natural language processing classification model could determine whether an input sentence was in French, Spanish, or Italian.",
                    "例如，某个自然语言处理分类模型可以确定输入的句子是法语、西班牙语还是意大利语。"
                ],
                [
                    "Compare with regression model.",
                    "请与回归模型进行比较。"
                ]
            ]
        },
        {
            "name": "classification threshold",
            "split": "test",
            "text": [
                [
                    "classification threshold",
                    "分类阈值"
                ],
                [
                    "A scalar-value criterion that is applied to a model's predicted score in order to separate the positive class from the negative class.",
                    "一种标量值条件，应用于模型预测的得分，旨在将正类别与负类别区分开。"
                ],
                [
                    "Used when mapping logistic regression results to binary classification.",
                    "将逻辑回归结果映射到二元分类时使用。"
                ],
                [
                    "For example, consider a logistic regression model that determines the probability of a given email message being spam.",
                    "以某个逻辑回归模型为例，该模型用于确定指定电子邮件是垃圾邮件的概率。"
                ],
                [
                    "If the classification threshold is 0.9, then logistic regression values above 0.9 are classified as spam and those below 0.9 are classified as not spam.",
                    "如果分类阈值为 0.9，那么逻辑回归值高于 0.9 的电子邮件将被归类为“垃圾邮件”，低于 0.9 的则被归类为“非垃圾邮件”。"
                ]
            ]
        },
        {
            "name": "clustering",
            "split": "test",
            "text": [
                [
                    "clustering",
                    "聚类"
                ],
                [
                    "Grouping related examples, particularly during unsupervised learning.",
                    "将关联的样本分成一组，一般用于非监督式学习。"
                ],
                [
                    "Once all the examples are grouped, a human can optionally supply meaning to each cluster.",
                    "在所有样本均分组完毕后，相关人员便可选择性地为每个聚类赋予含义。"
                ],
                [
                    "Many clustering algorithms exist.",
                    "聚类算法有很多。"
                ],
                [
                    "For example, the k-means algorithm clusters examples based on their proximity to a centroid, as in the following diagram:",
                    "例如，k-means 算法会基于样本与形心的接近程度聚类样本，如下图所示："
                ],
                [
                    "A human researcher could then review the clusters and, for example, label cluster 1 as \"dwarf trees\" and cluster 2 as \"full-size trees.\"",
                    "之后，研究人员便可查看这些聚类并进行其他操作，例如，将聚类 1 标记为“矮型树”，将聚类 2 标记为“全尺寸树”。"
                ],
                [
                    "As another example, consider a clustering algorithm based on an example's distance from a center point, illustrated as follows:",
                    "再举一个例子，例如基于样本与中心点距离的聚类算法，如下所示："
                ]
            ]
        },
        {
            "name": "convolutional neural network",
            "split": "test",
            "text": []
        },
        {
            "name": "cost",
            "split": "test",
            "text": []
        },
        {
            "name": "data augmentation",
            "split": "test",
            "text": []
        },
        {
            "name": "data set or dataset",
            "split": "test",
            "text": []
        },
        {
            "name": "decision boundary",
            "split": "test",
            "text": []
        },
        {
            "name": "dropout regularization",
            "split": "test",
            "text": []
        },
        {
            "name": "example",
            "split": "test",
            "text": []
        },
        {
            "name": "fairness constraint",
            "split": "test",
            "text": []
        },
        {
            "name": "few-shot learning",
            "split": "test",
            "text": []
        },
        {
            "name": "hyperparameter",
            "split": "test",
            "text": []
        },
        {
            "name": "implicit bias",
            "split": "test",
            "text": []
        },
        {
            "name": "items",
            "split": "test",
            "text": []
        },
        {
            "name": "label",
            "split": "test",
            "text": []
        },
        {
            "name": "landmarks",
            "split": "test",
            "text": []
        },
        {
            "name": "learning rate",
            "split": "test",
            "text": []
        },
        {
            "name": "loss surface",
            "split": "test",
            "text": []
        },
        {
            "name": "majority class",
            "split": "test",
            "text": []
        },
        {
            "name": "Markov decision process (MDP)",
            "split": "test",
            "text": []
        },
        {
            "name": "ML",
            "split": "test",
            "text": []
        },
        {
            "name": "MNIST",
            "split": "test",
            "text": []
        },
        {
            "name": "NaN trap",
            "split": "test",
            "text": []
        },
        {
            "name": "negative class",
            "split": "test",
            "text": []
        },
        {
            "name": "node (TensorFlow graph)",
            "split": "test",
            "text": []
        },
        {
            "name": "output layer",
            "split": "test",
            "text": []
        },
        {
            "name": "preprocessing",
            "split": "test",
            "text": []
        },
        {
            "name": "quantization",
            "split": "test",
            "text": []
        },
        {
            "name": "queue",
            "split": "test",
            "text": []
        },
        {
            "name": "representation",
            "split": "test",
            "text": []
        },
        {
            "name": "reward",
            "split": "test",
            "text": []
        },
        {
            "name": "Saver",
            "split": "test",
            "text": []
        },
        {
            "name": "step",
            "split": "test",
            "text": []
        },
        {
            "name": "timestep",
            "split": "test",
            "text": []
        },
        {
            "name": "trajectory",
            "split": "test",
            "text": []
        },
        {
            "name": "true positive rate (TPR)",
            "split": "test",
            "text": []
        },
        {
            "name": "wide model",
            "split": "test",
            "text": []
        }
    ]
}