Last updated: Apr 26, 2024
T-distributed Stochastic Neighbor Embedding (t-SNE) is a machine learning algorithm for
visualization. t-SNE charts model each high-dimensional object by a two-or-three dimensional point
in such a way that similar objects are modeled by nearby points and dissimilar objects are modeled
by distant points with high probability.
Creating a simple t-SNE chart
- In the Chart Type section, click the t-SNE
icon.
The canvas updates to display a t-SNE chart template.
- Set the Perplexity, Learning rate, and Maximum iterations values.
- Optional: Select a Color map variable.
- Click the Save visualization in the project control. Select Create a new asset or Append to existing asset. Provide a Visualization asset name, an optional description, and a chart name.
- Click Apply to save the visualization to the project. The new visualization asset is now available on the Assets tab.
Options
- Perplexity
- Sets a number that establishes an educated guess as to the number of close neighbors for each data point. The purpose is to balance the local and global aspects for your data.
- Learning rate
- This value affects the speed of learning by specifying the weight size changes at each iteration.
- Maximum iterations
- The maximum number of iterations to run.
- Color map
- Lists available color map variables. These variables use color progression, based on the range of values in the specified column, to represent themselves in the plot points. Color maps are also known as choropleth maps.
- Primary title
- The chart title.
- Subtitle
- The chart subtitle.
- Footnote
- The chart footnote.