uhh-lt path2vec issues - Githubissues

uhh-lt / path2vec

Learning to represent shortest paths and other graph-based measures of node similarities with graph embeddings

Apache License 2.0

32 stars 14 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

How to get the right embeddings for ambiguous words?

#28 mbarbouch closed 3 years ago
5
Training the model on the entire WordNet

#27 mbarbouch closed 3 years ago
19
Relation to Glove and word2vec

#26 alexanderpanchenko closed 5 years ago
8
Redraw the plot

#25 alexanderpanchenko closed 5 years ago
12
Obtain results for the SemEval 2013 (WSD)

#24 alexanderpanchenko closed 5 years ago
10
Compare to the GCN embeddings

#23 alexanderpanchenko closed 5 years ago
1
Recompute results for WSD for the best model

#22 alexanderpanchenko closed 5 years ago
0
Make the batch generator thread-safe

#21 akutuzov closed 5 years ago
4
specify the regularization coefficients from the command line

#20 alexanderpanchenko closed 5 years ago
0
Perform meta-parameter search using PyTorch implementation

#19 alexanderpanchenko closed 5 years ago
0
Results of regularized models on WordNet similarities

#18 akutuzov closed 5 years ago
11
Different Deepwalk results on static synset selection

#17 alexanderpanchenko closed 5 years ago
1
Rank the nearest neighbours for the regularization

#16 alexanderpanchenko closed 5 years ago
1
Add results of node2vec 600 to the paper to figure Figure 3

#15 alexanderpanchenko closed 5 years ago
20
Try the model without the similarities

#14 alexanderpanchenko closed 5 years ago
6
Try different dimensions of the embeddings

#13 alexanderpanchenko closed 5 years ago
0
Report the results for different values of beta/gamma

#12 alexanderpanchenko closed 6 years ago
1
Remove from the training dataset nodes with no neighbours and train the model with the neighbour regularizer (k=1) again

#11 alexanderpanchenko closed 5 years ago
0
Report results on all datasets with the regularization based on the adjacent nodes

#10 alexanderpanchenko closed 6 years ago
1
Report results on all datasets with L2 and other types of regularization

#9 alexanderpanchenko closed 5 years ago
5
Improvements in the embeddings script

#8 alexanderpanchenko closed 6 years ago
2
Compare to holographic graph embeddings

#7 alexanderpanchenko closed 5 years ago
34
Proof reading

#6 alexanderpanchenko closed 6 years ago
2
node2vec baseline

#5 alexanderpanchenko closed 6 years ago
38
graph construction code

#4 alexanderpanchenko closed 6 years ago
4
baseline based on the deepwalk

#3 alexanderpanchenko closed 6 years ago
37
Evaluation based on word sense disambiguation

#2 alexanderpanchenko closed 6 years ago
48
construction of a random graph

#1 alexanderpanchenko closed 6 years ago
0