machine-learning - 详尽的 channel /特征选择的降维

Question

我的数据由 16 个 channel x 128 个样本 x 400 个试验组成。我想在此数据集中执行详尽的 channel 选择。我应该在哪里应用 PCA？

unsortedChannelIndices = [1:16]
sortedChannelIndices = [];

%Option 1
reducedData = PCA(data, classIndeces)

for chIdx = 1:length(unsortedChannelIndices)

   for c=1:length(unsortedChannelIndices)
      thisChannel = unsortedChannelIndices(c)
      thisChannelSet = [sortedChannelIndices, thisChannel];

      %Option 1
      thisData = reducedData(thisChannelSet,:,:);

      %Option 2
      thisData = PCA(data(thisChannelSet, classIndeces)

      thisPerformance(c) = eval_perf(thisData);%crossvalidation
    end
    [performance(chIdx),best] = max(thisPerformance);
    sortedChannelIndices = [sortedChannelIndices,unsortedChannelIndices(best)];
    unsortedChannelIndices(best) =  [];
end

Answer 1

PCA 或任何降维技术应应用于将要分析的数据。如果我们想要评估与较少 channel (例如1:4)相对应的子集的性能，则应在此数据中应用任何降维技术(PCA(data([1:4),:,:)。因此，选项2是正确的选项。