lachhebo / pyclustertend Goto Github PK

I installed pyclustertend using pip install and also using conda install and many other ways, but is always giving the same error while running the code as
code=
import sys
import os.path
import os
import pyclustertend
from sklearn import datasets
from pyclustertend import vat
from sklearn.preprocessing import scale
X = scale(df3)
vat(X)

Error=

ModuleNotFoundError Traceback (most recent call last)
in
2 import os.path
3 import os
----> 4 import pyclustertend
5 from sklearn import datasets
6 from pyclustertend import vat

ModuleNotFoundError: No module named 'pyclustertend'

I tried many ways as given below:-

way 1#
(base) MacBook:~ pinky$ python -m pip install pyclustertend
Requirement already satisfied: pyclustertend in /opt/anaconda3/lib/python3.7/site-packages/pyclustertend-1.4.9-py3.7.egg (1.4.9)

way 2#
(base) MacBook:~ pinky$ pip3 install pyclustertend
Collecting pyclustertend
Using cached pyclustertend-1.4.9-py3-none-any.whl (9.8 kB)
Installing collected packages: pyclustertend
Successfully installed pyclustertend-1.4.9

way 3#
(base) MacBook:~ pinky$ conda install pyclustertend
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.

PackagesNotFoundError: The following packages are not available from current channels:

• pyclustertend

Current channels:

• https://conda.anaconda.org/conda-forge/osx-64
• https://conda.anaconda.org/conda-forge/noarch
• https://repo.anaconda.com/pkgs/main/osx-64
• https://repo.anaconda.com/pkgs/main/noarch
• https://repo.anaconda.com/pkgs/r/osx-64
• https://repo.anaconda.com/pkgs/r/noarch

To search for alternate channels that may provide the conda package you're
looking for, navigate to

https://anaconda.org

and use the search bar at the top of the page.

way4#
(base) MacBook:~ pinky$ pip install pyclustertend==1.4.9
Requirement already satisfied: pyclustertend==1.4.9 in /Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages (1.4.9)

But always getting error while ruuning in jupyter notebook as
ModuleNotFoundError: No module named 'pyclustertend'

Does anyone can help me to solve the problem, I am not very efficient in python, Thanks a lot in advance.

Hi, an error shows up during install, as below. Thanks.

pip seemed to fail to build package:
scikit-learn<0.25.0,>=0.24.0

Some possibly relevant errors from pip install:
error: subprocess-exited-with-error
ModuleNotFoundError: No module named 'numpy.distutils'
error: metadata-generation-failed
Error installing pyclustertend.

I can see that the usage condition is still contradictory to both Wikipedia as well as the reference in Pyclusterend docs.
Both Wikipedia and the reference indicate that a value closer to 1 indicate presence of clusters.

just installed pyclustertend 1.8.2 using pip as per instructions.
when trying to import the library i get the error above.
pyclustertend by default installs/upgrades scikit-learn 1.2.2. I downgraded to 1.1.2, but that didn't solve the problem.
Do you have any recommendations for how to make it work?
thanks a lot

Hello :)

I found some contradictory information on the hopkins test on https://pyclustertend.readthedocs.io/en/latest/

On the top it says:
If the test is positve, (an hopkins score which tends to 0) it means that clustering is useless for the dataset.

In the API section it says:
A score between 0 and 1, a score around 0.5 express no clusterability and a score tending to 0 express a high cluster tendency.

Firstly - great project! Thank you for your great work!

Coming to the issue, I use conda to manage my python dependencies, and it'd be really great if it were possible to release the package on conda-forge, so that it can easily be installed using conda.

Would love to know your thoughts on the same, thanks!

Hello,

First of all, thank you for implementiv VAT and iVAT! In the function compute_ivat_ordered_dissimilarity_matrix, the iVAT matrix is computed, but it is not symmetric in the end. Is is possible that there is some bug? I fixed it for now with the following code:
I changed
return re_ordered_matrix
to
for i in range(re_ordered_matrix.shape[0]):
for j in range(i):
re_ordered_matrix[j, i] = re_ordered_matrix[i, j]
return re_ordered_matrix

Sincere regards!

This package doesn't install on M1 Mac due version 0.24.2 of scikit-learn.
Could we bump the version of scikit-learn (0.24.2 is old and should be updated to something after the major release) ?

Hello.

I'd like to use your package to show the Hopkins test to my students but it, unfortunately, does not work with the latest scikit-learn. The problem is the old name for calinski_harabaz_score which has been removed from the latest sklearn.

I see that it's already fixed in the master so could you please release version 1.4.9 to PyPI?

Thanks a lot and have a nice day.