I'm trying to use the Predefined code in python 2.7 but whe I run the Setup and dataset loading code , I see an error
No module named pyLDAvis.sklearn ,
Do you know how to solve this?
Setup and dataset loading
First of all, let's load the libraries that we'll use.
import warnings # Disable some warnings
from dataiku import pandasutils as pdu
import pandas as pd, seaborn as sns
from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer
from sklearn.feature_extraction import text
from sklearn.decomposition import LatentDirichletAllocation,NMF
Populating the interactive namespace from numpy and matplotlib
ImportErrorTraceback (most recent call last)
<ipython-input-1-f0e756363047> in <module>()
10 from sklearn.decomposition import LatentDirichletAllocation,NMF
---> 11 import pyLDAvis.sklearn
ImportError: No module named pyLDAvis.sklearn
I fully recognize it's been over a year since you asked this question, but I just came across this and in the spirit of information sharing: any python error message saying "No module named..." indicates that either that package is misspelled, or the code environment used does not include this package.
The topic modeling plugin requires a python environment with NLP specific packages such as pyLDAvis, which are not included in DSS's built-in environment, which is likely what you were using.
You (or an admin, depending on your level of permissions) will need to first create a code environment that includes the packages imported at the top of the topic modeling notebook and then you'll want to change the notebook's code environment.