behrouz-babaki / minsizekmeans Goto Github PK

I have a dataset consisting of 1.3 million 256-dimensional vectors. Should I give it a try or is it a total waste of time?

Hi all,

I cloned this repo and made a fresh install of Anaconda (ver. 4.7.12), created a new environment using the following command:

conda create -n py35 python=3.5 anaconda

Followed by activating py35

conda activate py35

After making this environment, proceeded to run the code, throwing the following error:

Requested Python version (3.5) is not installed

Even though I confirm that the python version is correct:

python --version
Python 3.5.5

It won't run. Any advice would be great. I am currently running on Windows 7, and this is my full conda info list:

active environment : py35
active env location : C:\Users\Doug\Anaconda3\envs\py35
shell level : 2
user config file : C:\Users\Doug\.condarc
populated config files : C:\Users\Doug\.condarc
conda version : 4.7.12
conda-build version : 3.18.9
python version : 3.7.4.final.0
virtual packages :
base environment : C:\Users\Doug\Anaconda3 (writable)
channel URLs : https://conda.anaconda.org/conda-forge/win-64
https://conda.anaconda.org/conda-forge/noarch
https://repo.anaconda.com/pkgs/main/win-64
https://repo.anaconda.com/pkgs/main/noarch
https://repo.anaconda.com/pkgs/r/win-64
https://repo.anaconda.com/pkgs/r/noarch
https://repo.anaconda.com/pkgs/msys2/win-64
https://repo.anaconda.com/pkgs/msys2/noarch
package cache : C:\Users\Doug\Anaconda3\pkgs
C:\Users\Doug\.conda\pkgs
C:\Users\Doug\AppData\Local\conda\conda\pkgs
envs directories : C:\Users\Doug\Anaconda3\envs
C:\Users\Doug\.conda\envs
C:\Users\Doug\AppData\Local\conda\conda\envs
platform : win-64
user-agent : conda/4.7.12 requests/2.22.0 CPython/3.7.4 Windows/7 W
ows/6.1.7601
administrator : False
netrc file : None
offline mode : False

Hi,
I implemented the algorithm as a scikit classifier module so that is possible to place in scikit pipeline. Should I make a commit?

best = None best_clusters = None for i in range(args.NUM_ITER): clusters, centers = minsize_kmeans(data, args.k, args.min_size, args.max_size) if clusters: quality = compute_quality(data, clusters) if not best or (quality < best): best = quality best_clusters = clusters if best: if args.OUTFILE: with open(args.OUTFILE, 'w') as f: print('\n'.join(str(i) for i in clusters), file=f) else: print('cluster assignments:') for i in range(len(clusters)): print('%d: %d'%(i, clusters[i])) print('sum of squared distances: %.4f'%(best)) else: print('no clustering found')

Just a question, maybe I am wrong, but if you found a best cluster configuration, should not the variable best_clusters written to the file instead of the last cluster configuration? Regards Lukas

Hi,

How should I run on a CSV with multiple columns?

ex: I want to cluster data from a csv that contains:

1. Column One - Latitude
2. Column Two - Longitude

Thanks!