Comments (2)
@eliorc Thank you for the reply. I had done all that you describe and read several papers before asking the question. However, everything I've read seems to assume one large mostly connected graph esp. a social network or citations. I see my case as substantively different. I'll keep looking. Thanks again for your time.
from node2vec.
It's hard to say, but I'll try my best to explain the parameters' impact on your result.
Essentially you decide how you want to embed information from your graph into the embedding space using the different hyperparameters.
I'd say num_walk
is mostly for robustness, if your graphs are small enough I don't think you will gain from greater num_walk
- this only defines the number of walks you will created from each node. In turn, these random walks are the input data for a skip-gram algorithm (just like Word2Vec). So you are just creating more data.
Walk length, jointly with window_size
(which if you don't set, is set to 5 by gensim
), q
and p
define how these walks are constructed. It will be too much for me to explain here, but if you want to understand this better I'd suggest to
- Learn about the Word2Vec algorithm, and the role of window size in the algorithm
- Learn about
q
andp
(you can do so from looking at my PyData video in the README)
I think this will give you enough context to understand what you are looking for
from node2vec.
Related Issues (20)
- How can node2vec use GPU HOT 1
- too slow and stuck for some reason HOT 9
- reproducability HOT 2
- failing simple clustering of synthetic networks HOT 2
- Error in model creation node2vec HOT 1
- Cannot import Node2Vec HOT 5
- A faster implement HOT 1
- Train vs Inference methods HOT 6
- networkx>3.0 support? HOT 2
- There may be a small problem with the walking length. HOT 1
- Interpretation of weights HOT 2
- model save and loading examples HOT 2
- Are there any ways to initialize node vector using pretrained embeddings? HOT 1
- Crash when edges are not initialised with a weight attribute HOT 8
- can same temp directory be used for different graphs in for loop training? HOT 4
- _precompute_probabilities parallel execution HOT 2
- import error HOT 3
- segmentation fault HOT 5
- Error installing gensim in poetry env on mac m1 HOT 3
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 node2vec.