在 ND4J/DL4J 中重用归一化器

Question

我想知道在 ND4J/DL4J 中重用归一化器的正确方法是什么。目前，我将其保存如下：

final DataNormalization normalizer = new NormalizerStandardize();
normalizer.fit( trainingData );
normalizer.transform( trainingData );
normalizer.transform( testData );

try {
    final NormalizerSerializer normalizerSerializer = new NormalizerSerializer();
    normalizerSerializer.addStrategy( new StandardizeSerializerStrategy() );
    normalizerSerializer.write( normalizer, path );
} catch ( final IOException e ) {
    // ...
}

并通过以下方式加载：

try {
    final NormalizerSerializer normalizerSerializer = new NormalizerSerializer();
    normalizerSerializer.addStrategy( new StandardizeSerializerStrategy() );
    final DataNormalization normalizer = normalizerSerializer.restore( path );
} catch ( final Exception e ) { // Throws Exception instead of IOException.
    // ...
}

这样好吗？不幸的是，我无法在文档中找到更多信息。

Answer 1

这就是我的工作......

DataNormalization normalizer = new NormaizerStandardize();
normalizer.fit(trainingData);
normalizer.transform(trainingData);

保存

NormalizerSerializer saver = NormalizerSerializer.getDefaults();
File normalsFile = new File("fileName");
saver.write(normalizer,normalsFile);

恢复它

NormalizerSerializer loader = NormalizerSerializer.getDefaults();
DataNormalization restoredNormalizer = loader.restore(normalsFile);
restoredNormalizer.transform(testData);

ND4J Java 文档说.getDefaults() 获得了一个序列化器，配置了内置规范化器实现的策略。当您使用NormalizerStandardize 时，getDefaults() 提供了一种无需显式添加策略即可实现相同目的的简便方法。