Package org.nd4j.linalg.dataset.api.iterator.fetcher

Interface DataSetFetcher

    • Method Summary

      All Methods Instance Methods Abstract Methods 
      Modifier and Type Method Description
      int cursor()
      Direct access to a number represenative of iterating through a dataset
      void fetch​(int numExamples)
      Fetches the next dataset.
      boolean hasMore()
      Whether the dataset has more to load
      int inputColumns()
      The length of a feature vector for an individual example
      DataSet next()
      Returns the next data applyTransformToDestination
      void reset()
      Returns the fetcher back to the beginning of the dataset
      void setTopLevelDir​(File topLevelDir)
      For file download datasets this sets a top level directory outside the default
      File topLevelDir()
      The top level directory to download datasets from This will usually be the default dl4j resources directory at user home/.deeplearning4j
      int totalExamples()
      The total number of examples
      int totalOutcomes()
      The number of labels for a dataset

    • Method Detail

      • topLevelDir

        File topLevelDir()
        The top level directory to download datasets from This will usually be the default dl4j resources directory at user home/.deeplearning4j
        Returns:

      • setTopLevelDir

        void setTopLevelDir​(File topLevelDir)
        For file download datasets this sets a top level directory outside the default
        Parameters:
        topLevelDir - the top level directory to use

      • hasMore

        boolean hasMore()
        Whether the dataset has more to load
        Returns:
        whether the data applyTransformToDestination has more to load

      • next

        DataSet next()
        Returns the next data applyTransformToDestination
        Returns:
        the next dataset

      • fetch

        void fetch​(int numExamples)
        Fetches the next dataset. You need to call this to getFromOrigin a new dataset, otherwise next() just returns the last data applyTransformToDestination fetch
        Parameters:
        numExamples - the number of examples to fetch

      • totalOutcomes

        int totalOutcomes()
        The number of labels for a dataset
        Returns:
        the number of labels for a dataset

      • inputColumns

        int inputColumns()
        The length of a feature vector for an individual example
        Returns:
        the length of a feature vector for an individual example

      • totalExamples

        int totalExamples()
        The total number of examples
        Returns:
        the total number of examples

      • reset

        void reset()
        Returns the fetcher back to the beginning of the dataset

      • cursor

        int cursor()
        Direct access to a number represenative of iterating through a dataset
        Returns:
        a cursor similar to an index