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

Add temporal walk demo notebook #827

Closed
8 tasks
kjun9 opened this issue Feb 10, 2020 · 1 comment
Closed
8 tasks

Add temporal walk demo notebook #827

kjun9 opened this issue Feb 10, 2020 · 1 comment
Assignees
@kjun9
Labels
enhancement New feature or request experimental-blocker An issue that needs to be fixed before some `@experimental` feature can become non-experimental sg-library

Comments

@kjun9
Copy link
Contributor

kjun9 commented Feb 10, 2020

Description

We need a demo notebook to show that we can reproduce the findings in the reference paper for temporal walks

User Story

As a data scientist

I want to run a demo notebook for temporal walks

So that I can see its usefulness

Done Checklist

  • Produced code for required functionality
  • Tests written and coverage checked
  • Code review performed
  • Documentation on Google Docs (if applicable)
  • Documentation in repo
  • Version number reflects new status
  • CHANGELOG.md updated
  • Team demo
@kjun9 kjun9 added enhancement New feature or request sg-library labels Feb 10, 2020
@kjun9 kjun9 mentioned this issue Feb 10, 2020
Merged
kjun9 added a commit that referenced this issue Feb 11, 2020
@kjun9
Add temporal random walk as experimental (#787)
f8ca520 
Initial experimental implementation of temporal random walks.

Based on reference paper:

Giang Hoang Nguyen, John Boaz Lee, Ryan A. Rossi, Nesreen K. Ahmed, Eunyee
Koh, and Sungchul Kim. "Continuous-Time Dynamic Network Embeddings".
Proceedings of the 3rd International Workshop on Learning Representations
for Big Networks (WWW BigNet) 2018.

Part of #492 

Follow-ups #827 #828 #832
@kjun9 kjun9 self-assigned this Feb 18, 2020
@huonw huonw added the experimental-blocker An issue that needs to be fixed before some `@experimental` feature can become non-experimental label Feb 19, 2020
@kjun9 kjun9 mentioned this issue Mar 5, 2020
Closed
@kjun9
Copy link
Contributor Author

kjun9 commented Mar 9, 2020

Current version of this isn't producing results that match the paper, but the performance still looks good - to move forward, we could just make a note of the observed differences for now.

@kjun9 kjun9 mentioned this issue Mar 10, 2020
Merged
kjun9 added a commit that referenced this issue Mar 11, 2020
@kjun9
Add enron employees dataset (#1058)
e07ed44 
This adds a dataset loader for the `ia-enron-employees` dataset called 
`IAEnronEmployees`, so that it can be used for the CTDNE demo notebook #827 

This also introduces a new parameter for dataset loaders called 
`url_archive_contains_directory` which can be used to allow for downloading 
archives that do not contain a directory for its files.

Dataset loaders must now always be used with keyword arguments - this 
allows for related arguments to be grouped together regardless of which 
ones have default values or not.
kjun9 added a commit that referenced this issue Mar 19, 2020
@kjun9
Add CTDNE (time-based random walk) demo (#1056)
659e012 
This adds a demo notebook for the reference paper:

Nguyen, Giang & Lee, John & Rossi, Ryan & Ahmed, Nesreen & Koh, Eunyee & 
Kim, Sungchul. (2018). Continuous-Time Dynamic Network Embeddings.

It uses the `TemporalRandomWalk` class to perform link prediction on the 
`ia-enron-employees` dataset, and compares the results against the baseline 
method of running biased random walks which does not use the temporal 
information on edges.

See #827
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@kjun9 kjun9

Labels
enhancement New feature or request experimental-blocker An issue that needs to be fixed before some `@experimental` feature can become non-experimental sg-library
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@huonw @timpitman @kjun9