Closed wehlutyk closed 6 years ago
Currently running in my session on grunch
, using https://github.com/ixxi-dante/nw2vec/blob/master/projects/scale/blogcatalog.py with dim_ξ = 10
.
Training is done, must look at the results now.
Results are in 1f4c4106daf838033dc4e7dae8c7d0ed1f980d35, see the projects/scale/blogcatalog-dim_ξ=10-results.ipynb notebook.
Highlights (see in the figures below, extracted from the notebook):
Now:
I don't think looking at higher dimensions will give anything else. Instead, there are (at least) two other points to check:
And combinations of those two. If all that fails to explain the bad adjacency reconstruction, then the answers will be found by working on the behaviour project: #30 and #32 mainly.
Closing this issue in favour of all the above.
To see if that's the limiting factor in why the embeddings currently don't look good at all. (See here.)