Denali is a tool for visualizing complex and possibly high-dimensional data intuitively and efficiently. The central concept used by denali is that of the landscape metaphor: a 2-d "mountainous terrain" embedded in 3-d which preserves important features of the data and presents them to the user in an intuitive fashion. This software package includes tools for computing, visualizing, and manipulating landscape metaphors.

Can I use denali to visualize my data?

Broadly-speaking, there are two pre-requisites for visualizing your data with denali:

  1. Each data point must have an associated scalar value, such as a probability, score, or cost.
  2. You must have some method of extracting tree-like structure from your data.

Included with denali are tools for satisfying the second prerequisite for many types of data that may not appear, at first glance, to exhibit tree-like structure.

For example, suppose we have drawn many samples from a probability distribution. Can we attempt to understand the structure of the distribution by visualizing the samples with denali? For each sample we have an associated probability, so the first prerequisite is clearly met. However our data is in the form of a point cloud, not a tree, and so it appears that the second requirement is not satisfied. In fact, we can extract tree-like structure from this point cloud in a very natural way using the concept of the contour tree. The contour tree captures the important topological characteristics of a scalar function, and, as its name implies, it exhibits tree-like structure. Hence our second requirement is met, and we can visualize the probability distribution using denali.

Denali includes tools for quickly and easily computing contour trees, making it simply and straightforward to visualize many different types of data as landscape metaphors.

What is included with denali?

Several tools are provided alongside denali. ctree is tool for computing contour trees from point clouds. pydenali is a package of Python utilities for interacting with denali and ctree.

Using this documentation

The suggested method of using this documentation is as follows:

Next: Installation