Wolfram Language & System 11.0 (2016)|Legacy Documentation

This is documentation for an earlier version of the Wolfram Language.View current documentation (Version 11.2)

WLNet (.wlnet)

  • Used for storing trained and untrained neural nets.
  • Binary format based on the "HDF5" format.
  • Developed at Wolfram Research.

BackgroundBackground

Import and ExportImport and Export

  • Import["file.wlnet"] imports the net stored in the specified file, returning it as a NetGraph or similar construct.
  • Export["file.wlnet",expr] exports the NetGraph or similar construct expr to a file.

ExamplesExamplesopen allclose all

Basic Examples  (2)Basic Examples  (2)

Define a simple neural network:

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

Export this network to the WLNet format:

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

Import the network:

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

Define an initialized graph network:

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

Export the network along with its initialized parameters:

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

Import the network:

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

Apply the imported network to data:

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

This is equivalent to evaluating the original network:

In[5]:=
Click for copyable input
Out[5]=
Introduced in 2016
(11.0)