Keras-fit_generator中的class_weight vs sample_weights [英] Keras - class_weight vs sample_weights in the fit_generator

问题描述

在Keras中(使用TensorFlow作为后端)，我正在构建一个模型，该模型正在处理具有高度不平衡类(标签)的巨大数据集.为了能够运行训练过程，我创建了一个生成器，该生成器将大块数据馈送到fit_generator.

In Keras (using TensorFlow as a backend) I am building a model which is working with a huge dataset that is having highly imbalanced classes (labels). To be able to run the training process, I created a generator which feeds chunks of data to the fit_generator.

根据 fit_generator 的文档，生成器的输出可以是元组(inputs, targets)或元组(inputs, targets, sample_weights).考虑到这一点，这里有几个问题:

According to the documentation for the fit_generator, the output of the generator can either be the tuple (inputs, targets) or the tuple (inputs, targets, sample_weights). Having that in mind, here are a few questions:

1. 我的理解是 class_weight考虑整个数据集所有类的权重，而 sample_weights考虑每个单独块的所有类的权重 由生成器创建.那是对的吗?如果没有，有人可以详细说明吗?
2. 是否有必要将class_weight分别赋予fit_generator，然后再赋予sample_weights作为每个块的输出?如果是，那为什么呢?如果不是，那么哪个最好给?
3. 如果应该为每个块提供sample_weights，如果特定块中缺少某些类，如何映射权重?让我举个例子吧.在我的整体数据集中，我有7种可能的类(标签).因为这些类高度不平衡，所以当我创建较小的数据块作为fit_generator的输出时，某些块会丢失特定的块.我应该如何为这些块创建sample_weights?

1. My understanding is that the class_weight regards the weights of all classes for the entire dataset whereas the sample_weights regards the weights of all classes for each individual chunk created by the generator. Is that correct? If not, can someone elaborate on the matter?
2. Is it necessary to give both the class_weight to the fit_generator and then the sample_weights as an output for each chunk? If yes, then why? If not then which one is better to give?
3. If I should give the sample_weights for each chunk, how do I map the weights if some of the classes are missing from a specific chunk? Let me give an example. In my overall dataset, I have 7 possible classes (labels). Because these classes are highly imbalanced, when I create smaller chunks of data as an output from the fit_generator, some of the classes are missing from the specific chunk. How should I create the sample_weights for these chunks?

推荐答案

我的理解是，class_weight考虑了所有对象的权重 整个数据集的类，而sample_weights考虑 所有类别的权重由 发电机.那是对的吗?如果没有，有人可以详细说明 有关系吗?

My understanding is that the class_weight regards the weights of all classes for the entire dataset whereas the sample_weights regards the weights of all classes for each individual chunk created by the generator. Is that correct? If not, can someone elaborate on the matter?

class_weight影响目标函数计算中每个类别的相对权重.顾名思义，sample_weights允许进一步控制属于同一类别的样品 的相对权重.

class_weight affects the relative weight of each class in the calculation of the objective function. sample_weights, as the name suggests, allows further control of the relative weight of samples that belong to the same class.

是否必须同时将class_weight赋予fit_generator和 那么sample_weights作为每个块的输出?如果是，那为什么呢? 如果不是，那么哪个更适合?

Is it necessary to give both the class_weight to the fit_generator and then the sample_weights as an output for each chunk? If yes, then why? If not then which one is better to give?

这取决于您的应用程序.在对高度偏斜的数据集进行训练时，班级权重非常有用；例如，用于检测欺诈交易的分类器.当您对批次中的样品没有相同的信心时，样品重量非常有用.一个常见的示例是对具有不确定性的变量进行回归分析.

It depends on your application. Class weights are useful when training on highly skewed data sets; for example, a classifier to detect fraudulent transactions. Sample weights are useful when you don't have equal confidence in the samples in your batch. A common example is performing regression on measurements with variable uncertainty.

如果我应该为每个块都提供sample_weights，我该如何映射 如果某个特定块中缺少某些类，则权重是多少?让 我举个例子.在我的整体数据集中，我有7个可能的类 (标签).因为这些类高度不平衡，所以当我创建 较小的数据块作为fit_generator的输出，其中一些 特定块中缺少类.我应该如何创建 这些块的sample_weights?

If I should give the sample_weights for each chunk, how do I map the weights if some of the classes are missing from a specific chunk? Let me give an example. In my overall dataset, I have 7 possible classes (labels). Because these classes are highly imbalanced, when I create smaller chunks of data as an output from the fit_generator, some of the classes are missing from the specific chunk. How should I create the sample_weights for these chunks?

这不是问题. sample_weights是在每个样本的基础上定义的，并且与类无关.因此，文档指出(inputs, targets, sample_weights)的长​​度应相同.

This is not an issue. sample_weights is defined on a per-sample basis and is independent from the class. For this reason, the documentation states that (inputs, targets, sample_weights) should be the same length.

功能 _weighted_masked_objective >有一个示例sample_weights正在应用.

The function _weighted_masked_objective in engine/training.py has an example of sample_weights are being applied.

这篇关于Keras-fit_generator中的class_weight vs sample_weights的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！