Scilab Home Page | Wiki | Bug Tracker | Forge | Mailing List Archives | Scilab Online Help | File Exchange
ATOMS : Lolimot details
Please login or create an account

Lolimot

A fast neural network - LOcal LInear MOdel Tree
(1018 downloads for this version - 14742 downloads for all versions)
Details
Version
2.1
Author
Yann COLLETTE
Owner Organization
Home
Maintainers
Administrator Atoms
Yann Collette
Samuel Gougeon
License
Creation Date
December 10, 2018
Source created on
Scilab 6.0.x
Binaries available on
Scilab 5.5.x:
Windows 64-bit Windows 32-bit Linux 64-bit Linux 32-bit MacOSX
Scilab 6.0.x:
Windows 64-bit Windows 32-bit Linux 64-bit Linux 32-bit MacOSX
Scilab 6.1.x:
Windows 64-bit Windows 32-bit Linux 64-bit Linux 32-bit MacOSX
Install command
--> atomsInstall("lolimot")
Description
            
The Lolimot model (LOcal LInear MOdel Tree) is a model from the neural network
family. It is able to learn every kind of non-linear relationship.
Compared to a classical neural network, the learning phase of a LOLIMOT model is
quite fast and deterministic.
Once a model is learnt, you have an easy access to analytical derivatives of
your model through a call to the function estim_der_lolimot.

FUNCTIONS
---------
lolimot_estim — Computes the Lolimot estimation for a given vector
lolimot_estim_der — Analytical Lolimot derivative of a learnt model
lolimot_estim_vec — Computes the Lolimot estimation for a given vector
lolimot_export_der_model — Exports the partial derivative of a Lolimot model
into a .sci file
lolimot_export_model — Exports a Lolimot model to a .sci file
lolimot_learn — Learns a Lolimot model for a given inputs/output data set
lolimot_learn_valid — Learns a Lolimot model for a given inputs/output data
set and using a validation data set
lolimot_plot_part — Plot the Lolimot cuts for a 2D model
lolimot_read — Read a Lolimot model
lolimot_update — The function update the linear models of a given Lolimot
model for a given learning data set
lolimot_write — Saves a Lolimot model in a file

DOCUMENTS
---------
User manual: https://atoms.scilab.org/toolboxes/lolimot/2.1/files/Lolimot_21_manual.pdf
Changelog: https://atoms.scilab.org/toolboxes/lolimot/2.1/files/changelog.txt

WARNING: Most of functions have been renamed, with the 'lolimot_' prefix
            
Files (6)
[1.52 kB]
Miscellaneous file
Changes 2.0 -> 2.1
------------------
* Portage to Scilab 6
  - eval => evstr
  - getf => exec
  - printf => mprintf
  - MSDOS => getos
  - builder.sce

* functions renamed:
  - estim_lolimot       => lolimot_estim
  - estim_vec_lolimot   => lolimot_estim_vec
  - estim_vec_lolimot   => lolimot_estim_der
  - learn_lolimot       => lolimot_learn
  - learn_valid_lolimot => lolimot_learn_valid
  - read_lolimot        => lolimot_read
  - write_lolimot       => lolimot_write
  - update_lolimot      => lolimot_update
  - plot_lolimot_part   => lolimot_plot_part
  - export_model        => lolimot_export_model
  - export_der_model    => lolimot_export_der_model

* Help pages
 - Encoding converted to UTF-8.
 - See also: short descriptions added.

* Bugs fixed:
 - in learn_lolimot(), modelOut definition as a mlist() had a wrong syntax.
 - in estim_der_lolimot(), dU and dV initialized to [] yielded some
   operation +[] warnings and subsequent errors.
 - in export_model() and export_der_model(), "\%eps" instead of "%%eps"
   in mfprintf() yielded some errors.
 - The demo wrote a file in the current directory instead of in TMPDIR.
 - The demo yielded some warning about plot().
 - Demo : Operation +[] issues fixed.
   
* Improvements:
 - _export_model() and export_der_model() could record data only in the
   current directory.

[165.43 kB]
Source code archive

[204.69 kB]
OS-independent binary for Scilab 5.5.x

[233.30 kB]
OS-independent binary for Scilab 6.0.x

[233.30 kB]
OS-independent binary for Scilab 6.1.x

[163.88 kB]
Miscellaneous file
User manual
News (0)
Comments (0)
Leave a comment
You must register and log in before leaving a comment.
Email notifications
Send me email when this toolbox has changes, new files or a new release.
You must register and log in before setting up notifications.