DimensionReducerFunction

DimensionReducerFunction[]

represents a function generated by DimensionReduction that projects data onto a lower-dimensional approximating manifold.

Details and Options

  • DimensionReducerFunction works like Function.
  • DimensionReducerFunction[][example] projects the single example example onto the approximating manifold.
  • DimensionReducerFunction[][{example1,example2,}] projects all the examplei.
  • DimensionReducerFunction[][data,prop] gives the specified property of the dimension reduction associated with data.
  • Possible properties include:
  • "ReducedVectors"reduction of the vectors given
    "OriginalData"deduced original examples given reduced vectors
    "ReconstructedData"reconstruction by reduction and inversion
    "ImputedData"missing values replaced by imputed ones

Examples

open allclose all

Basic Examples  (3)

Generate a DimensionReducerFunction from a list of vectors:

In[1]:=
Click for copyable input
Out[1]=

Use this reducer function on a new vector:

In[11]:=
Click for copyable input
Out[11]=

Use this reducer function on a list of new vectors:

In[12]:=
Click for copyable input
Out[12]=

The vectors can also contain missing values:

In[13]:=
Click for copyable input
Out[13]=

Generate a DimensionReducerFunction from a list of vectors:

In[1]:=
Click for copyable input
Out[1]=

Use this reducer function on a list of new vectors:

In[2]:=
Click for copyable input
In[3]:=
Click for copyable input
Out[3]=

Try to reconstruct the original vectors from the reduced ones:

In[4]:=
Click for copyable input
Out[4]=

The reconstructed vectors can be directly obtained from the original vectors:

In[5]:=
Click for copyable input
Out[5]=

Generate a dimension reducer on a mixed-type dataset:

In[1]:=
Click for copyable input
Out[1]=

Reduce the dimension of a new example:

In[2]:=
Click for copyable input
Out[2]=

Scope  (1)

See Also

DimensionReduction  DimensionReduce  ClassifierFunction  PredictorFunction

Introduced in 2015
(10.1)
| Updated in 2017
(11.1)