Comments (2)
MaartenGr
commented on July 17, 2024
1
The models are not interchangeable between 0.5.0 and 0.4.3. This means that loading a model from 0.4.3 into 0.5.0 is unlikely to work, especially since parameters differ and the API changes frequently between major updates.
If you want to calculate the probabilities in v0.5, you should set calculate_probabilities to True. It is set to False as a default because it is computationally an inefficient process in HDBSCAN which can result in a significant increase in running times.
Good catch on the wording in the documentation! That seems to be left-over from 0.4.3 when low_memory and calculate_probabilities were still combined. I'll make sure to chance it in v0.6.0 which is currently in the works.
from bertopic.
makkasu
commented on July 17, 2024
Thanks very much for the swift response! Didn't spot the calculate_probabilities thing, that seems to have done the trick. Cheers
from bertopic.
Related Issues (20)
- TerminatedWorkerError:A worker process managed by the executor was unexpectedly terminated. This could be caused by a segmentation fault while calling the function or by an excessive memory usage causing the Operating System to kill the worker. The exit codes of the workers are {EXIT(1)} HOT 3
- TypeError:'NoneType' object is not subscriptable while calling topic_model.hierarchical_topics HOT 5
- Can't update model name when use notebook HOT 4
- Scikit-learn's HDBSCAN Implementation
- Issues with visualizations on loaded models. HOT 1
- (Zero-shot Topic Modeling) TypeError: object of type 'numpy.float64' has no len() HOT 1
- Additional representations did not update with topic reduction HOT 5
- [Guided Topic Modeling] ValueError: setting an array element with a sequence. HOT 6
- Array mismatch when try to fit new data HOT 1
- Does Bertopic support custom keyword extractor? HOT 5
- Why do I lose Names assigned by zero-shot after applying outlier reduction? HOT 1
- get_topic() with KeyBERTInspired? HOT 1
- Mismatch between old OpenAI API in bertopic/backend/_openai and current OpenAI (v1.33.0) HOT 1
- OpenAI Embedding HOT 1
- Potential bug with the PartOfSpeech class due to lower case matching HOT 1
- Chosen represented Topic HOT 5
- merge_topics from a model loaded from safe tensors HOT 1
- Add test coverage around OpenAI(BaseRepresentation) HOT 3
- Can't import bertopic in pycharm but runs fine in Anaconda prompt console HOT 3
- Adding representation model does not change topic_model.get_document_info results() HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from bertopic.