Data

Tools

Indexes

Applications

Experiments

Open Source Science

cancer_research_gnn

https://github.com/oniani/cancer_research_gnn

Science Score: 41.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ○
    .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ✓
    Committers with academic emails
    1 of 3 committers (33.3%) from academic institutions
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (1.4%) to scientific vocabulary
Last synced: 10 months ago · JSON representation ·

Repository

Basic Info
  • Host: GitHub
  • Owner: oniani
  • Language: Python
  • Default Branch: master
  • Size: 135 KB
Statistics
  • Stars: 0
  • Watchers: 2
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created almost 6 years ago · Last pushed almost 6 years ago
Metadata Files
Readme Citation

README.md

Results

GraphSAGE

Mean Aggregator

StatisticsHyperparameters

| Statistic | Value | | --------- | ------------------ | | Accuracy | 0.9012345679012346 | | Precision | 0.9261904761904762 | | Recall | 0.9102607709750566 | | F-Score | 0.9147043432757718 |

| Hyperparameter | Value | | --------------------------- | ------------- | | Dropout probability | 0.25 | | Learning rate | 1e-2 (0.01) | | Number of training epochs | 800 | | Number of hidden gcn units | 16 | | Number of hidden gcn layers | 1 | | Weight for L2 loss | 5e-4 (0.0005) | | Aggregator type | mean |

GCN Aggregator

StatisticsHyperparameters
| Statistic | Value | | --------- | -------------------- | | Accuracy | 0.2222222222222222 | | Precision | 0.031746031746031744 | | Recall | 0.14285714285714285 | | F-Score | 0.051948051948051945 | | Hyperparameter | Value | | --------------------------- | ------------- | | Dropout probability | 0.25 | | Learning rate | 1e-1 (0.1) | | Number of training epochs | 800 | | Number of hidden gcn units | 2 | | Number of hidden gcn layers | 1 | | Weight for L2 loss | 5e-4 (0.0005) | | Aggregator type | gcn |

MoNet

StatisticsHyperparameters
| Statistic | Value | | --------- | -------------------- | | Accuracy | 0.2222222222222222 | | Precision | 0.031746031746031744 | | Recall | 0.14285714285714285 | | F-Score | 0.051948051948051945 | | Hyperparameter | Value | | -------------------------------------------------------------------- | ------------- | | Dropout probability | 0.25 | | Learning rate | 1e-1 (0.1) | | Number of training epochs | 800 | | Number of hidden gcn units | 2 | | Number of hidden gcn layers | 1 | | Pseudo coordinate dimensions in GMMConv, 2 for cora and 3 for pubmed | 2 | | Number of kernels in GMMConv layer | 3 | | Weight for L2 loss | 5e-4 (0.0005) |

GAT

StatisticsHyperparameters
| Statistic | Value | | --------- | ----- | | Accuracy | 0.753 | | Precision | 0.726 | | Recall | 0.684 | | F-Score | 0.697 | | Hyperparameter | Value | | ------------------------------------------ | ----------- | | Number of training epochs | 1000 | | Number of hidden attention heads | 4 | | Uumber of output attention heads | 1 | | Number of hidden layers | 1 | | Number of hidden units | 200 | | Use residual connection | False | | Input feature dropout | 0 | | Attention dropout | 0 | | Learning rate | 1e-2 (0.01) | | Weight decay | 0 | | The negative slope of leaky relu | 0.2 | | Indicates whether to use early stop or not | False | | Skip re-evaluate the validation set | False |

GCN

StatisticsHyperparameters
| Statistic | Value | | --------- | ------------------ | | Accuracy | 0.8024691358024691 | | Precision | 0.8550170307357392 | | Recall | 0.786734693877551 | | F-Score | 0.8035058227176454 | | Hyperparameter | Value | | --------------------------- | ----------- | | Dropout probability | 0 | | Learning rate | 1e-2 (0.01) | | Number of training epochs | 4000 | | Number of hidden gcn units | 500 | | Number of hidden gcn layers | 1 | | Weight for L2 loss | 0 |

APPNP

StatisticsHyperparameters
| Statistic | Value | | --------- | ------------------ | | Accuracy | 0.8888888888888888 | | Precision | 0.9251082251082252 | | Recall | 0.8943877551020407 | | F-Score | 0.9049666689418242 | | Hyperparameter | Value | | --------------------------- | ------------- | | Input feature dropout | 0.25 | | Edge propagation dropout | 0.5 | | Learning rate | 1e-1 (0.1) | | Number of training epochs | 800 | | Hidden unit sizes for appnp | [64] | | Number of propagation steps | 10 | | Teleport Probability | 0.4 | | Weight for L2 loss | 5e-4 (0.0005) |

GIN

StatisticsHyperparameters
| Statistic | Value | | --------- | ------------------ | | Accuracy | 0.8271604938271605 | | Precision | 0.8205627705627706 | | Recall | 0.8091836734693878 | | F-Score | 0.8045525902668759 | | Hyperparameter | Value | | ------------------------- | ---------------- | | Extra args | [16, 1, 0, True] | | Learning rate | 1e-2 (0.01) | | Weight decay | 5e-6 (0.000005) | | Number of training epochs | 800 |

TAGCN

StatisticsHyperparameters
| Statistic | Value | | --------- | ------------------ | | Accuracy | 0.9012345679012346 | | Precision | 0.9070381998953428 | | Recall | 0.9102607709750566 | | F-Score | 0.9041060526774812 | | Hyperparameter | Value | | ------------------------- | -------------------- | | Extra args | [16, 1, F.relu, 0.5] | | Learning rate | 1e-2 (0.01) | | Weight decay | 5e-4 (0.0005) | | Number of training epochs | 800 |

SGC

StatisticsHyperparameters
| Statistic | Value | | --------- | ------------------ | | Accuracy | 0.8271604938271605 | | Precision | 0.8362389490209041 | | Recall | 0.8315759637188209 | | F-Score | 0.8240298807695121 | | Hyperparameter | Value | | ------------------------- | ---------------- | | Extra args | [None, 1, False] | | Learning rate | 1e-1 (0.1) | | Weight decay | 0 | | Number of training epochs | 4000 |

AGNN

StatisticsHyperparameters
| Statistic | Value | | --------- | ------------------ | | Accuracy | 0.8765432098765432 | | Precision | 0.88992673992674 | | Recall | 0.8888321995464853 | | F-Score | 0.8850179383028748 | | Hyperparameter | Value | | ------------------------- | ------------------------ | | Extra args | [100, 1, 1.0, True, 0.1] | | Learning rate | 1e-1 (0.1) | | Weight decay | 0 | | Number of training epochs | 200 |

ChebNet

StatisticsHyperparameters
| Statistic | Value | | --------- | ------------------ | | Accuracy | 0.9012345679012346 | | Precision | 0.9022735409953455 | | Recall | 0.9201814058956915 | | F-Score | 0.9073651359365645 | | Hyperparameter | Value | | ------------------------- | ---------------- | | Extra args | [32, 1, 2, True] | | Learning rate | 1e-2 (0.001) | | Weight decay | 5e-4 (0.0005) | | Number of training epochs | 800 |

Feature Engineering

sh python generate_data.py --num_classes=6 python feature_engineering.py

Owner

  • Name: David Oniani
  • Login: oniani
  • Kind: user
  • Location: Arlington, VA, USA

Citation (citation_network/README.md) 

# Node Classification on Citation Networks

This example shows how to use modules defined in `dgl.nn.pytorch.conv` to do node classification on
citation network datasets.

## Datasets

- Cora
- Citeseer
- Pubmed

## Models

- GCN: [Semi-Supervised Classification with Graph Convolutional Networks](https://arxiv.org/pdf/1609.02907)
- GAT: [Graph Attention Networks](https://arxiv.org/abs/1710.10903)
- GraphSAGE [Inductive Representation Learning on Large Graphs](https://cs.stanford.edu/people/jure/pubs/graphsage-nips17.pdf)
- APPNP: [Predict then Propagate: Graph Neural Networks meet Personalized PageRank](https://arxiv.org/pdf/1810.05997)
- GIN: [How Powerful are Graph Neural Networks?](https://arxiv.org/abs/1810.00826)
- TAGCN: [Topology Adaptive Graph Convolutional Networks](https://arxiv.org/abs/1710.10370)
- SGC: [Simplifying Graph Convolutional Networks](https://arxiv.org/abs/1902.07153)
- AGNN: [Attention-based Graph Neural Network for Semi-supervised Learning](https://arxiv.org/pdf/1803.03735.pdf)
- ChebNet: [Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering](https://arxiv.org/abs/1606.09375)

## Usage

```
python run.py [--gpu] --model MODEL_NAME --dataset DATASET_NAME [--self-loop]
```

The hyperparameters might not be the optimal, you could specify them manually in `conf.py`.

GitHub Events

Total
Last Year

Committers

Last synced: 12 months ago

All Time
  • Total Commits: 49
  • Total Committers: 3
  • Avg Commits per committer: 16.333
  • Development Distribution Score (DDS): 0.49
Past Year
  • Commits: 0
  • Committers: 0
  • Avg Commits per committer: 0.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
David Oniani;Oniani.David@mayo.edu;30005663 m****2@i****u 25
David Oniani o****d@g****m 18
sfc s****2@g****m 6
Committer Domains (Top 20 + Academic)
infodev8.mayo.edu: 1

Issues and Pull Requests

Last synced: 12 months ago

All Time
  • Total issues: 0
  • Total pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Total issue authors: 0
  • Total pull request authors: 0
  • Average comments per issue: 0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 0
  • Pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Issue authors: 0
  • Pull request authors: 0
  • Average comments per issue: 0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
Pull Request Authors
Top Labels
Issue Labels
Pull Request Labels