search

MIND-Lab / OCTIS

OCTIS: Comparing Topic Models is Simple! A python package to optimize and evaluate topic models (accepted at EACL2021 demo track)
MIT License
734 stars 106 forks source link

Dashboard - python error #82

Open fatmas1982 opened 1 year ago

fatmas1982 commented 1 year ago

Description

runtime error

What I Did

from octis.models.LDA import LDA 
cause this error 
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
[<ipython-input-54-5a8917c9d1ca>](https://localhost:8080/#) in <module>
----> 1 import octis.models.LDA

6 frames
/usr/local/lib/python3.8/dist-packages/gensim/_matutils.pyx in init gensim._matutils()

ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 96 from C header, got 88 from PyObject
silviatti commented 1 year ago

Hello! Could you provide more details on your environment? It's difficult for us to replicate the error otherwise. Thanks!

Silvia

Chrisma-98 commented 1 year ago

Hello! Could you provide more details on your environment? It's difficult for us to replicate the error otherwise. Thanks!

Silvia

Hi! Silvia. When I was training CTM on my computer, I encountered "Unable to allocate 26.0 GiB for an array with shape (55776, 62456) and data type int64". I thought it might be a GPU memory problem, so I used colab. I encountered the same problem when I ran the project on colab.

What I Did Based on stackoverflow's search results, I tried to reinstall numpy(1.19.1 and 1.20.3) and gensim (3.5.0, 3.6.0 and 4.1.2). It doesn't work. image