To get a visual read on cohesiveness, we can apply UMAP to reduce the image classification embeddings all the way down to 2 dimensions. Samples of thumbnails from 2 clusters are shown below:ģ. Observe how Resnet captures color information very effectively (see all the blue shoes in the second cluster) but sometimes fails to encode shape information (see the first cluster). Almost as simple: we can apply the last hidden layer of an off-the-shelf pretrained image classification network (Resnet 18), and evaluate embedding quality by clustering them. We can see that while the first PC distinguishes between interpretable types of avatars, the twelfth is too broad to be meaningful.Ģ. To evaluate the “quality” of the reduction, we visualize thumbnails on the extremes of the principal components (PCs).
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |