t-SNE Reduction Node

T

AI/ML/Reduction

t-Distributed Stochastic Neighbor Embedding. Projects high-dimensional vectors into 2-3 dimensions for visualization and writes the embedding back into the source table. t-SNE is transductive, so it produces no reusable model.

fit_tsneml

Inputs6

Outputs1

Security exposure6/10

Packageml

Ratings

Scores range from 0 to 10. Higher values mean more impact, exposure, or operational weight.

SecurityAttack surface and exposure impact.

6/10Medium

PrivacyPotential sensitivity of processed data.

6/10Medium

PerformanceRuntime or resource pressure.

2/10Low

GovernancePolicy, audit, or compliance impact.

5/10Medium

ReliabilityOperational stability considerations.

6/10Medium

CostExternal or compute cost impact.

3/10Low

Input Pins

6

Input

Execution

exec_in

Execution trigger that begins the t-SNE embedding

Data Source

String

source

Choose which backend supplies the data

Default Database

Database

Embedding Size

Integer

embedding_size

Dimensionality of the embedding. Must not exceed the width of the input vectors; values above 3 require the exact gradient (Approx Threshold = 0).

Default 2

Range 1 to 3

Perplexity

Float

perplexity

Effective number of neighbors per point (typically 5-50). t-SNE requires 3 * perplexity <= rows - 1, so small tables need a small perplexity.

Default 30

Range 1 to 200

Approx Threshold

Float

approx_threshold

Barnes-Hut theta. 0 runs the exact O(n^2) gradient, larger values approximate distant points by their cell centroid and run faster.

Default 0.5

Range 0 to 2

Max Iterations

Integer

max_iter

Number of gradient descent iterations. Fewer iterations finish sooner but may leave the embedding unconverged.

Default 1000

Range 50 to 10000

Output Pins

1

Done

Execution

exec_out

Activated once the t-SNE embedding has been written back

Node Info

Internal name

fit_tsne

Category

AI/ML/Reduction

Version

2