Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[WIP] Super nodes #17

Merged
merged 56 commits into from
Sep 16, 2022
Merged

[WIP] Super nodes #17

merged 56 commits into from
Sep 16, 2022

Conversation

vict0rsch
Copy link
Collaborator

@vict0rsch vict0rsch commented Jun 15, 2022 •
edited by AlexDuvalinho
Loading

  • verify importance of tag 0 nodes in prediction
  • remove all tag 0 nodes
  • one super node per graph
  • one supernode per atom type
  • one supernode per atom type with min distance
  • positional embeddings

AlexDuvalinho
AlexDuvalinho reviewed Jun 21, 2022
View reviewed changes
get_data_sample.py
sn_pos = [b.pos[sub_nodes[i]].mean(0) for i in range(batch_size)]
# target relaxed position is the mean of the super node's aggregates
# (per batch)
sn_pos_relaxed = [
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Same as sn_pos since tag-0 nodes are fixed. No need to compute it.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

indeed ; very small overhead though

@RolnickLab RolnickLab deleted a comment from vict0rsch Jun 22, 2022
@AlexDuvalinho
first draft one supernode per node type
c10d62f
@AlexDuvalinho
In process creation of supernode per atom type --Co-authored-by: Vict…
2197d1c 
…or Schmidt
@AlexDuvalinho
second draft supernode extensions
ca01b36
@AlexDuvalinho
refine phys_embeddings: remove col + add ionization energy
1a69889
@AlexDuvalinho
proper implem of one-supernode-per-graph graph rewiring
4ba7c77
@AlexDuvalinho
working supernodes implem
a34036b
@AlexDuvalinho
refactor one-supernode-per-atom-type inside dist fct, not finished
e58ff99
@AlexDuvalinho
continue refactor one-supernode-per-atom-type inside dist fct
f600539
@AlexDuvalinho
good version of one-supernode-per-graph
d6e683e
@AlexDuvalinho
correct typo in names
a3824cc
@AlexDuvalinho
supernode with min. subnode distance to other atoms
9df1170
@AlexDuvalinho
one tensor not on correct device
5825615
@AlexDuvalinho
clean graph rewiring
b7bdf8e
@AlexDuvalinho
relaxed_pos correction
73dc673
@AlexDuvalinho
add positional encoding for schnet, supernode extensions only
5f5dc1e
@AlexDuvalinho
positional encoding for all methods
8071768
@AlexDuvalinho
fix one-supernode-per-graph issue (part 1)
91ffd9d
@AlexDuvalinho
fix pe + one-supernode-per-graph issue
629ee8f
@AlexDuvalinho AlexDuvalinho marked this pull request as ready for review July 13, 2022 15:13
@vict0rsch vict0rsch merged commit 4ca9b47 into main Sep 16, 2022
@vict0rsch vict0rsch deleted the supernode branch September 16, 2022 17:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@AlexDuvalinho AlexDuvalinho AlexDuvalinho left review comments

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@vict0rsch @AlexDuvalinho