Stack Overflow
Stack Overflow

Questions tagged [som]

A self-organizing map (SOM) is a type of artificial neural network that is trained using unsupervised learning to produce a map reprensenting the input space of the training samples.

A self-organizing map (SOM) is a type of artificial neural network that is trained using unsupervised learning to produce a map representing the input space of the training samples. Also, some supervised and semi-supervised SOMs exist.

187 questions
1
vote
1 answer

Plot class probability by neuron in self organizing maps

I found a nice tutorial of self organizing map clustering in R in which, it is explained how to display your input data in the unit space (see below). In order to set up some rules for the labeling, I would like to compute the probability of each…
WAF
  • 1,141
  • 20
  • 44
1
vote
3 answers

SOM Algorithm Matlab HELP

I'm trying to pass a txt file to som_read_data from a GUI......i created a function that takes a txt file from the GUI and then passes it to som_read_data..but i'm getting some errors...here are a list of some of the errors.....any one with…
Tim
  • 61
  • 1
  • 1
  • 12
1
vote
0 answers

How is unit classification created in som, using kohonen package in R

Code used: system.time(som_model <- som(train_matrix, grid=som_grid, rlen= p, alpha=c(0.05,0.001), n.hood = "circular", …
Parth Khare
  • 11
  • 2
1
vote
1 answer

Normalize data with R

I have a dataframe called data which contains 5 columns and approximately 181 rows. I'm trying to run some algorithm on this dataframe, but I have to do some pre-processing beforehand and normalize the columns to have zero and 1. I am using R and…
user3603831
  • 87
  • 1
  • 6
1
vote
1 answer

How to implement Kohonen maps (SOMs) with non-scalar input data/attributes

In general, the implementation of a Kohonen map / SOM algorithm with only real-valued vectors is a relatively trivial task. I wonder though how to implement such an algorithm for non-real-valued (i.e. non-scalar) attributes such as text strings, due…
sunside
  • 8,069
  • 9
  • 51
  • 74
1
vote
1 answer

Reading C code written for R packages. Where to find definitions for RANDIN, UNIF, EPS, etc?

I am interested in precisely understanding C code for R packages. But I have come across code such as RANDIN, RANDOUT, UNIF, EPS etc (are they macros?) for which I don't know where to find the implementations/definitions. Where can I find the code…
FXQuantTrader
  • 6,821
  • 3
  • 36
  • 67
1
vote
2 answers

U-Matrix for Rectangular Maps

I have read about the U-Matrix in many places including this site. The best explanation of a U-Matrix is found here in this site with the explanation why there is so little correct information (the original paper is not at all useful) about how the…
Synex
  • 207
  • 6
  • 18
1
vote
1 answer

Self Organizing Map "errors"

As we know from K-Means after the sample data are clustered in N clusters (each cluster has a centroid vector) not all the data are clustered in the clusters that they belong to! I mean that some data vectors may be clustered in wrong clusters. This…
Drill
  • 169
  • 2
  • 3
  • 12
1
vote
2 answers

Appropriate neighborhood size for SOM

I'm doing a program that can cluster numeric data using Kohonen Self-Organizing Maps and I'm trying to make it as generic as possible. So, how do I know the appropriate initial size of the neighborhood in proportion to the number of items (number of…
libzz
  • 589
  • 2
  • 10
  • 29
1
vote
2 answers

Similarity algorithm (mathematics) of sampled signals

Let's say I have sampled some signals and constucted a vector of the samples for each. What is the most efficent way to calculate (dis)similarity of those vectors? Note that offset of the sampling must not count, for instance sample-vectors of sin…
Simo Erkinheimo
  • 1,347
  • 9
  • 17
0
votes
0 answers

Plotting an array as a point in 2D space in Python?

I have some lists with a large length and I want to plot them in 2D (like a scatter plot). The thing is, I need to maintain their topology / preserve their distance when I do this mapping. If the distance(A,B) > distance(A,C), it should stay that…
user21629433
  • 1
0
votes
0 answers

How do you know which is the closest class example(observation) to a class in Self Organizing Map?

For example there are two classes C1 and C2, distance1 and distance2, The rule tells which class is: If distance1 > distance2 than class C1 (Node1) else class C2 (Node2) Here are some data examples: distance1 distance2 class 5.315 …
darieem
  • 1
  • 4
0
votes
1 answer

Debugger only shows Python bytecode when debugging SOMPY library functions

I am trying to set up a self-organizing map (SOM) with the python library SOMPY right now. I made some changes to the original code since I would like to implement an alternative method for the computation of the best-matching unit (BMU). However I…
Yeggers
  • 11
  • 4
0
votes
0 answers

Plotting a static hexagonal topology using MiniSom with a color scheme

I'm using the MiniSom package to make a Self Organizing Map of my data, in one of their examples on GitHub they plot a nice static hexagonal topology visualising the distance between the neighbouring neurons using a color based on an array with…
Bas R
  • 175
  • 7
0
votes
0 answers

Clustering using Quicksom package

I am trying to cluster similar data points using quicksom package which is GPU based SOM using Pytorch. All the functions are running smoothly except when I am using som.predict() function that calculates quantization errors and topographic errors.…
user13497521
  • 1
  • 1
1 2 3
…
12 13