Document Type

Article

Abstract

Manifolk is a tool to visualize the output of dimensionality reduction algorithms like t-SNE, PCA etc. One of this tool’s main uses is that it de-clutters graphs by plotting data points pertaining to a subset of labels. The subset of labels to be plotted can be selected using the provided checkboxes. A case study on data from a publicly available action recognition dataset like UCF101 shows how this tool can help find outliers. With the rise in self-supervised methods for training deep neural networks, this tool helps researchers better visualize the embeddings learned by the model.

Publication Date

7-2-2021

Language

English

License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Share

COinS