RAPIDS
    Preparing search index...

    Class FittedUMAP

    Hierarchy (View Summary)

    Index

    Accessors

    a b embeddings init initialAlpha learningRate localConnectivity minDist nComponents negativeSampleRate nEpochs nNeighbors randomState repulsionStrength setOpMixRatio spread targetMetric targetNNeighbors targetWeight transformQueueSize verbosity

    Methods

    fitArray fitDataFrame fitSeries getEmbeddings getUMAPParams refineArray refineDataFrame refineSeries setUMAPParams transformArray transformDataFrame transformSeries

    Accessors

    • get a(): number

      Returns number

    • get b(): number

      Returns number

    • get embeddings(): Embeddings

      Returns Embeddings

      Embeddings in low-dimensional space in float32 format, which can be converted to any of the following types: DataFrame, Series, DeviceBuffer.

      // returns DataFrame<{[K extends number]: Series<Float32>}>
      embeddings.asDataFrame();
      // returns Series<Float32>
      embeddings.asSeries();
      // returns rmm.DeviceBuffer
      embeddings.asDeviceBuffer();
    • get init(): number

      Returns number

    • get initialAlpha(): number

      Returns number

    • get learningRate(): number

      Returns number

    • get localConnectivity(): number

      Returns number

    • get minDist(): number

      Returns number

    • get nComponents(): number

      Returns number

    • get negativeSampleRate(): number

      Returns number

    • get nEpochs(): number

      Returns number

    • get nNeighbors(): number

      Returns number

    • get randomState(): number

      Returns number

    • get repulsionStrength(): number

      Returns number

    • get setOpMixRatio(): number

      Returns number

    • get spread(): number

      Returns number

    • get targetMetric(): string

      Returns string

    • get targetNNeighbors(): number

      Returns number

    • get targetWeight(): number

      Returns number

    • get transformQueueSize(): number

      Returns number

    • get verbosity(): string

      Returns string

    Methods

    • Fit features into an embedded space.

      Parameters

      • features: MemoryData

        array containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

        // For a sample dataset of colors, with properties r,g and b:
        features = [
          ...Object.values({ r: xx1, g: xx2, b: xx3 }),
          ...Object.values({ r: xx4, g: xx5, b: xx6 }),
        ] // [xx1, xx2, xx3, xx4, xx5, xx6]
      • Optionaltarget: MemoryData | null

        array containing target values

        // For a sample dataset of colors, with properties r,g and b:
        target = [color1, color2] // len(target) = nFeatures
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1, y1, x2, y2...]

      Returns FittedUMAP

      FittedUMAP object with updated embeddings

      This method will automatically convert the inputs to float32

    • Fit features into an embedded space.

      Parameters

      • features: DeviceBuffer

        array containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

        // For a sample dataset of colors, with properties r,g and b:
        features = [
          ...Object.values({ r: xx1, g: xx2, b: xx3 }),
          ...Object.values({ r: xx4, g: xx5, b: xx6 }),
        ] // [xx1, xx2, xx3, xx4, xx5, xx6]
      • Optionaltarget: DeviceBuffer | null

        array containing target values

        // For a sample dataset of colors, with properties r,g and b:
        target = [color1, color2] // len(target) = nFeatures
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1, y1, x2, y2...]

      Returns FittedUMAP

      FittedUMAP object with updated embeddings

      This method will automatically convert the inputs to float32

    • Fit features into an embedded space.

      Type Parameters

      Parameters

      • features: T

        array containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

        // For a sample dataset of colors, with properties r,g and b:
        features = [
          ...Object.values({ r: xx1, g: xx2, b: xx3 }),
          ...Object.values({ r: xx4, g: xx5, b: xx6 }),
        ] // [xx1, xx2, xx3, xx4, xx5, xx6]
      • Optionaltarget: R | null

        array containing target values

        // For a sample dataset of colors, with properties r,g and b:
        target = [color1, color2] // len(target) = nFeatures
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1, y1, x2, y2...]

      Returns FittedUMAP

      FittedUMAP object with updated embeddings

      This method will automatically convert the inputs to float32

    • Fit features into an embedded space.

      Parameters

      • features: (number | bigint | null | undefined)[]

        array containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

        // For a sample dataset of colors, with properties r,g and b:
        features = [
          ...Object.values({ r: xx1, g: xx2, b: xx3 }),
          ...Object.values({ r: xx4, g: xx5, b: xx6 }),
        ] // [xx1, xx2, xx3, xx4, xx5, xx6]
      • Optionaltarget: (number | bigint | null | undefined)[] | null

        array containing target values

        // For a sample dataset of colors, with properties r,g and b:
        target = [color1, color2] // len(target) = nFeatures
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1, y1, x2, y2...]

      Returns FittedUMAP

      FittedUMAP object with updated embeddings

      This method will automatically convert the inputs to float32

    • Fit features into an embedded space

      Type Parameters

      Parameters

      • features: DataFrame<{ [P in string]: T }>

        Dense or sparse matrix containing floats or doubles. Acceptable dense formats: cuDF DataFrame

      • Optionaltarget: Series<R>

        cuDF Series containing target values

        // For a sample dataset of colors, with properties r,g and b:
        target = [color1, color2] // len(target) = nFeatures

      Returns FittedUMAP

      FittedUMAP object with updated embeddings

      This method will automatically convert the inputs to float32

    • Fit features into an embedded space

      Type Parameters

      Parameters

      • features: T

        cuDF Series containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

      • Optionaltarget: R | null

        cuDF Series containing target values

        // For a sample dataset of colors, with properties r,g and b:
        target = [color1, color2] // len(target) = nFeatures
      • nFeatures: number = 1

        number of properties in the input features, if features is of the format [x1,y1,x2,y2...]

      Returns FittedUMAP

      FittedUMAP object with updated embeddings

      This method will automatically convert the inputs to float32

    • Returns Embeddings

      Embeddings in low-dimensional space in dtype format, which can be converted to any of the following types: DataFrame, Series, DeviceBuffer.

      // returns DataFrame<{[K extends number]: Series<Float32>}>
      getEmbeddings(new Float64).asDataFrame();
      // returns Series<Float32>
      getEmbeddings(new Int32).asSeries();
      // returns rmm.DeviceBuffer
      getEmbeddings(new UInt32).asDeviceBuffer();

    • Refine features into existing embedded space as base

      Parameters

      • features: MemoryData

        array containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

        // For a sample dataset of colors, with properties r,g and b:
        features = [
          ...Object.values({ r: xx1, g: xx2, b: xx3 }),
          ...Object.values({ r: xx4, g: xx5, b: xx6 }),
        ] // [xx1, xx2, xx3, xx4, xx5, xx6]
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1, y1, x2, y2...]

      Returns void

      This method will automatically convert the inputs to float32

    • Refine features into existing embedded space as base

      Parameters

      • features: DeviceBuffer

        array containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

        // For a sample dataset of colors, with properties r,g and b:
        features = [
          ...Object.values({ r: xx1, g: xx2, b: xx3 }),
          ...Object.values({ r: xx4, g: xx5, b: xx6 }),
        ] // [xx1, xx2, xx3, xx4, xx5, xx6]
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1, y1, x2, y2...]

      Returns void

      This method will automatically convert the inputs to float32

    • Refine features into existing embedded space as base

      Type Parameters

      Parameters

      • features: T

        array containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

        // For a sample dataset of colors, with properties r,g and b:
        features = [
          ...Object.values({ r: xx1, g: xx2, b: xx3 }),
          ...Object.values({ r: xx4, g: xx5, b: xx6 }),
        ] // [xx1, xx2, xx3, xx4, xx5, xx6]
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1, y1, x2, y2...]

      Returns void

      This method will automatically convert the inputs to float32

    • Refine features into existing embedded space as base

      Parameters

      • features: (number | bigint | null | undefined)[]

        array containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

        // For a sample dataset of colors, with properties r,g and b:
        features = [
          ...Object.values({ r: xx1, g: xx2, b: xx3 }),
          ...Object.values({ r: xx4, g: xx5, b: xx6 }),
        ] // [xx1, xx2, xx3, xx4, xx5, xx6]
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1, y1, x2, y2...]

      Returns void

      This method will automatically convert the inputs to float32

    • Refine features into existing embedded space as base

      Type Parameters

      Parameters

      • features: DataFrame<{ [P in string]: T }>

        Dense or sparse matrix containing floats or doubles. Acceptable dense formats: cuDF DataFrame

      Returns void

      This method will automatically convert the inputs to float32

    • Refine features into existing embedded space as base

      Type Parameters

      Parameters

      • features: T

        cuDF Series containing floats or doubles in the format [x1, y1, z1, x2, y2, z2...] for features x, y & z.

      • nFeatures: number = 1

        number of properties in the input features, if features is of the format [x1,y1,x2,y2...]

      Returns void

      This method will automatically convert the inputs to float32

    • Transform features into the existing embedded space and return that transformed output.

      Parameters

      • features: MemoryData
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1,y1,x2,y2...]

      Returns Embeddings

      Transformed features into the existing embedded space and return an Embeddings instance which can be converted to any of the following types: DataFrame, Series, DeviceBuffer.

      transformArray(...).asDataFrame(); // returns DataFrame<{number: Series<Numeric>}>
      transformArray(...).asSeries(); // returns Series<Numeric>
      transformArray(...).asDeviceBuffer(); //returns rmm.DeviceBuffer

      This method will automatically convert the inputs to float32

    • Transform features into the existing embedded space and return that transformed output.

      Parameters

      • features: DeviceBuffer
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1,y1,x2,y2...]

      Returns Embeddings

      Transformed features into the existing embedded space and return an Embeddings instance which can be converted to any of the following types: DataFrame, Series, DeviceBuffer.

      transformArray(...).asDataFrame(); // returns DataFrame<{number: Series<Numeric>}>
      transformArray(...).asSeries(); // returns Series<Numeric>
      transformArray(...).asDeviceBuffer(); //returns rmm.DeviceBuffer

      This method will automatically convert the inputs to float32

    • Transform features into the existing embedded space and return that transformed output.

      Type Parameters

      Parameters

      • features: T
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1,y1,x2,y2...]

      Returns Embeddings

      Transformed features into the existing embedded space and return an Embeddings instance which can be converted to any of the following types: DataFrame, Series, DeviceBuffer.

      transformArray(...).asDataFrame(); // returns DataFrame<{number: Series<Numeric>}>
      transformArray(...).asSeries(); // returns Series<Numeric>
      transformArray(...).asDeviceBuffer(); //returns rmm.DeviceBuffer

      This method will automatically convert the inputs to float32

    • Transform features into the existing embedded space and return that transformed output.

      Parameters

      • features: (number | bigint | null | undefined)[]
      • OptionalnFeatures: number

        number of properties in the input features, if features is of the format [x1,y1,x2,y2...]

      Returns Embeddings

      Transformed features into the existing embedded space and return an Embeddings instance which can be converted to any of the following types: DataFrame, Series, DeviceBuffer.

      transformArray(...).asDataFrame(); // returns DataFrame<{number: Series<Numeric>}>
      transformArray(...).asSeries(); // returns Series<Numeric>
      transformArray(...).asDeviceBuffer(); //returns rmm.DeviceBuffer

      This method will automatically convert the inputs to float32

    • Transform features into the existing embedded space and return that transformed output.

      Type Parameters

      Parameters

      Returns Embeddings

      Transformed features into the existing embedded space and return an Embeddings instance which can be converted to any of the following types: DataFrame, Series, DeviceBuffer

      transformDataFrame(...).asDataFrame(); // returns DataFrame<{number: Series<Numeric>}>
      transformDataFrame(...).asSeries(); // returns Series<Numeric>
      transformDataFrame(...).asDeviceBuffer(); //returns rmm.DeviceBuffer

      This method will automatically convert the inputs to float32

    • Transform features into the existing embedded space and return that transformed output.

      Type Parameters

      Parameters

      • features: T
      • nFeatures: number = 1

        number of properties in the input features, if features is of the format [x1,y1,x2,y2...]

      Returns Embeddings

      Transformed features into the existing embedded space and return an Embeddings instancewhich can be converted to any of the following types: DataFrame, Series, DeviceBuffer.

      transformSeries(...).asDataFrame(); // returns DataFrame<{number: Series<Numeric>}>
      transformSeries(...).asSeries(); // returns Series<Numeric>
      transformSeries(...).asDeviceBuffer(); //returns rmm.DeviceBuffer

      This method will automatically convert the inputs to float32

    Settings

    Member Visibility

    On This Page

    Accessors
    Methods