Tensor.UMAP¶
| Name | Mandatory | Description | Default | Type |
|---|---|---|---|---|
⬅️ Input |
The input of the shard, if any | Tensor |
||
Output ➡️ |
The resulting output of the shard | Tensor |
||
OutputDims |
No | Number of dimensions in the output. Default is 3. | 3 |
IntVar(Int)None |
Neighbors |
No | Number of neighbors to consider for each point. Default is 15. | 15 |
IntVar(Int)None |
MinDist |
No | Minimum distance between points. Default is 0.1. | 0.1 |
FloatVar(Float)None |
Epochs |
No | Number of epochs for optimization. Default is 200. | 200 |
IntVar(Int)None |
LearningRate |
No | Learning rate for optimization. Default is 1.0. | 1 |
FloatVar(Float)None |
Metric |
No | Distance metric to use. Options: 'euclidean', 'cosine'. Default is 'euclidean'. | euclidean |
StringNone |
Init |
No | Initialization method. Options: 'spectral', 'random'. Default is 'spectral'. | spectral |
StringNone |
Seed |
No | Random seed for reproducibility. Default is 42. | 42 |
IntVar(Int)None |
Clusters |
No | Number of clusters to detect. Default is 0 (auto-detect). | 0 |
IntVar(Int)None |
Reduces dimensionality using UMAP (Uniform Manifold Approximation and Projection) algorithm.