https://github.com/deric/clueminer
interactive clustering platform
https://github.com/deric/clueminer
clustering clustering-algorithm
Last synced: about 2 months ago
JSON representation
interactive clustering platform
- Host: GitHub
- URL: https://github.com/deric/clueminer
- Owner: deric
- License: other
- Created: 2012-11-14T16:56:09.000Z (over 12 years ago)
- Default Branch: master
- Last Pushed: 2023-07-10T08:29:32.000Z (almost 2 years ago)
- Last Synced: 2025-04-23T04:08:20.875Z (about 2 months ago)
- Topics: clustering, clustering-algorithm
- Language: Java
- Homepage:
- Size: 18.7 MB
- Stars: 15
- Watchers: 5
- Forks: 11
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: license.txt
Awesome Lists containing this project
README
# Clueminer
[](https://travis-ci.org/deric/clueminer)
Clueminer is a platform for interactive data-mining with special focus on clustering algorithms.
## Building latest development version
Prerequisites:
* git
* Java 8 or newer
* Maven
* on Debian/Ubuntu Linux:
apt install maven libjhdf5 git
git clone git://github.com/deric/clueminer.git
cd clueminer
git submodule init
git submodule update
mvn clean install
* Once build you can run Clueminer via Maven
cd modules/application
mvn nbm:cluster-app nbm:run-platform
* or use generated bin file
bash modules/application/target/clueminer/bin/clueminer
### From NetBeans
1. open the clueminer directory (which is a maven module) -- this module is called `clueminer-parent`
2. from dependent modules of `clueminer-parent` open `clueminer-app`
3. execute "Build with dependencies"
4. now you can run the main application (run module `clueminer-app`)
## How to increase heap size
In `application/src/main/resources/clueminer.conf` adjust Java options:
default_options="--branding clueminer -J-Xms24m -J-Xmx2048m"
Note: increasing heap size `J-Xmx` to values bigger than is your actual physical RAM will
cause serious preformance issues!
On Unix systems you can find out your memory size with this command:
```bash
echo $(( $(awk '/MemTotal/{print $2}' /proc/meminfo) >> 10 ))m
```
## Forking
1. click on fork button on github
2. add upstream repository
```
$ git remote add upstream https://github.com/deric/clueminer.git
```
3. from time to time merge with upstream
```
$ git fetch upstream
$ git checkout master
$ git merge upstream/master
```
## Benchmarks
Benchmarks of clustering algorithms are located in `modules/clustering-benchmark`
In order to run benchmarks build an asssembly with Maven:
```
$ cd modules/clustering-benchmark
$ mvn assembly:assembly
```
and run benchmarks (might be computationally expensive):
``
$ java -jar target/*-jar-with-dependencies.jar
``
## R support
In order to enable R code execution JRI is needed.
Debian/Ubuntu:
```
apt install r-cran-rjava
```
## HDF support
Library `libjhdf5` is needed to be present on library path.
Debian/Ubuntu:
```
apt install libjhdf5-jni
```
## OpenGL support
For OpenGL visualizations you'll need native extensions
### Debian/Ubuntu
jzy3d is using jogl library which has native bindings:
sudo apt-get install libjogl-java
## R support
Currently there are two possibilities how to execute R code from Java:
* TCP connection to `RServe`
* load dynamic library `libjri.so`
* for Debian: `apt install r-cran-rjava`, make sure `R_HOME` is set
* update classpath `export CLASSPATH=.:/usr/lib/R/site-library/rJava/jri/`