dyf #3
Open
dyf #3
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Jupyter notebook for arvix_data_loading_pipeline
Add some explanation for extracting normal embedding from he saved graph. And mention the problem may happen processing dataset mag
…00 nodes). This sampled mag nodes have more nodes, but less connected compared to arxiv dataset.
This is the sampled graph mag dataset. It contains paper node and paper citation edge only and "feat" is the 128 embedding of title and abstract provided by the dataset, "_ID" is the node_ID, "y" is the class labels. its structure is Node data: dict_keys(['year', 'feat', '_ID', '_TYPE', 'y']) Edge data: dict_keys(['reltype', '_ID', '_TYPE'])
Here is the description image of arxiv dataset, mag dataset and sampled mag dataset.
This is model trained on graphsage on sampled mag dataset. I split the 2w data into train, validatino and test, seperated by year of paper like 2013, 2015. The accuracy (47% for training 100 epoches) is higher than the 2005.00687v7 arxiv paper, it provided only 31.53% . Maybe my sampled dataset is better connected, so it is easier to predict the node label. This is the first version, training is really faster than I thought, it only takes minutes, I will finish the other part later.
recommend to change the name for better understanding
Upload two notebook for MLP training, and Result evaluation
This jupyter file contains the data processing progress that all we need. Including sampling and combination. Also, it contains Graphsage training on the combined graph.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.