umap
Visualize and reduce high-dimensional data into lower dimensions using uniform manifold approximation and projection (UMAP) for enhanced analysis, tailored for single-cell RNA sequencing workflows.
Instructions
Uniform Manifold Approximation and Projection (UMAP) for visualization
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| a | No | Parameter controlling the embedding. | |
| alpha | No | Initial learning rate for the embedding optimization. | |
| b | No | Parameter controlling the embedding. | |
| gamma | No | Weighting applied to negative samples. | |
| init_pos | No | How to initialize the low dimensional embedding. | spectral |
| maxiter | No | Number of iterations (epochs) of the optimization. | |
| method | No | Implementation to use ('umap' or 'rapids'). | umap |
| min_dist | No | Minimum distance between embedded points. | |
| n_components | No | Number of dimensions of the embedding. | |
| negative_sample_rate | No | Number of negative samples per positive sample. | |
| neighbors_key | No | Key for neighbors settings in .uns. | |
| random_state | No | Random seed for reproducibility. | |
| spread | No | Scale of embedded points. |
Input Schema (JSON Schema)
{
"description": "Input schema for the UMAP dimensionality reduction tool.",
"properties": {
"a": {
"anyOf": [
{
"exclusiveMinimum": 0,
"type": "number"
},
{
"type": "null"
}
],
"default": null,
"description": "Parameter controlling the embedding.",
"title": "A"
},
"alpha": {
"default": 1,
"description": "Initial learning rate for the embedding optimization.",
"exclusiveMinimum": 0,
"title": "Alpha",
"type": "number"
},
"b": {
"anyOf": [
{
"exclusiveMinimum": 0,
"type": "number"
},
{
"type": "null"
}
],
"default": null,
"description": "Parameter controlling the embedding.",
"title": "B"
},
"gamma": {
"default": 1,
"description": "Weighting applied to negative samples.",
"exclusiveMinimum": 0,
"title": "Gamma",
"type": "number"
},
"init_pos": {
"default": "spectral",
"description": "How to initialize the low dimensional embedding.",
"title": "Init Pos",
"type": "string"
},
"maxiter": {
"anyOf": [
{
"exclusiveMinimum": 0,
"type": "integer"
},
{
"type": "null"
}
],
"default": null,
"description": "Number of iterations (epochs) of the optimization.",
"title": "Maxiter"
},
"method": {
"default": "umap",
"description": "Implementation to use ('umap' or 'rapids').",
"title": "Method",
"type": "string"
},
"min_dist": {
"default": 0.5,
"description": "Minimum distance between embedded points.",
"exclusiveMinimum": 0,
"title": "Min Dist",
"type": "number"
},
"n_components": {
"default": 2,
"description": "Number of dimensions of the embedding.",
"exclusiveMinimum": 0,
"title": "N Components",
"type": "integer"
},
"negative_sample_rate": {
"default": 5,
"description": "Number of negative samples per positive sample.",
"exclusiveMinimum": 0,
"title": "Negative Sample Rate",
"type": "integer"
},
"neighbors_key": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"description": "Key for neighbors settings in .uns.",
"title": "Neighbors Key"
},
"random_state": {
"default": 0,
"description": "Random seed for reproducibility.",
"title": "Random State",
"type": "integer"
},
"spread": {
"default": 1,
"description": "Scale of embedded points.",
"exclusiveMinimum": 0,
"title": "Spread",
"type": "number"
}
},
"title": "UMAPModel",
"type": "object"
}