{"id":20324381,"url":"https://github.com/oniani/cancer_research_gnn","last_synced_at":"2025-07-14T08:32:44.069Z","repository":{"id":112155265,"uuid":"281739981","full_name":"oniani/cancer_research_gnn","owner":"oniani","description":null,"archived":false,"fork":false,"pushed_at":"2020-09-09T00:50:43.000Z","size":138,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-07-05T11:51:42.459Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/oniani.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":"citation_network/README.md","codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2020-07-22T17:20:45.000Z","updated_at":"2021-03-30T02:53:53.000Z","dependencies_parsed_at":null,"dependency_job_id":"9cd69fcb-b12c-434f-aa38-98211bccff6d","html_url":"https://github.com/oniani/cancer_research_gnn","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/oniani/cancer_research_gnn","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oniani%2Fcancer_research_gnn","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oniani%2Fcancer_research_gnn/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oniani%2Fcancer_research_gnn/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oniani%2Fcancer_research_gnn/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/oniani","download_url":"https://codeload.github.com/oniani/cancer_research_gnn/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/oniani%2Fcancer_research_gnn/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":265262640,"owners_count":23736439,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-14T19:33:47.459Z","updated_at":"2025-07-14T08:32:43.486Z","avatar_url":"https://github.com/oniani.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Results\n\n- [GraphSAGE](#graphsage)\n  - [Mean Aggregator](#mean-aggregator)\n  - [GCN Aggregator](#gcn-aggregator)\n- [MoNet](#monet)\n- [GAT](#gat)\n- [GCN](#gcn)\n- [APPNP](#appnp)\n- [GIN](#gin)\n- [TAGCN](#tagcn)\n- [SGC](#sgc)\n- [AGNN](#agnn)\n- [ChebNet](#chebnet)\n\n## GraphSAGE\n\n### Mean Aggregator\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\u003ctable\u003e\u003c/table\u003e\n\n| Statistic | Value              |\n| --------- | ------------------ |\n| Accuracy  | 0.9012345679012346 |\n| Precision | 0.9261904761904762 |\n| Recall    | 0.9102607709750566 |\n| F-Score   | 0.9147043432757718 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter              | Value         |\n| --------------------------- | ------------- |\n| Dropout probability         | 0.25          |\n| Learning rate               | 1e-2 (0.01)   |\n| Number of training epochs   | 800           |\n| Number of hidden gcn units  | 16            |\n| Number of hidden gcn layers | 1             |\n| Weight for L2 loss          | 5e-4 (0.0005) |\n| Aggregator type             | mean          |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n### GCN Aggregator\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value                |\n| --------- | -------------------- |\n| Accuracy  | 0.2222222222222222   |\n| Precision | 0.031746031746031744 |\n| Recall    | 0.14285714285714285  |\n| F-Score   | 0.051948051948051945 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter              | Value         |\n| --------------------------- | ------------- |\n| Dropout probability         | 0.25          |\n| Learning rate               | 1e-1 (0.1)    |\n| Number of training epochs   | 800           |\n| Number of hidden gcn units  | 2             |\n| Number of hidden gcn layers | 1             |\n| Weight for L2 loss          | 5e-4 (0.0005) |\n| Aggregator type             | gcn           |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## MoNet\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value                |\n| --------- | -------------------- |\n| Accuracy  | 0.2222222222222222   |\n| Precision | 0.031746031746031744 |\n| Recall    | 0.14285714285714285  |\n| F-Score   | 0.051948051948051945 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter                                                       | Value         |\n| -------------------------------------------------------------------- | ------------- |\n| Dropout probability                                                  | 0.25          |\n| Learning rate                                                        | 1e-1 (0.1)    |\n| Number of training epochs                                            | 800           |\n| Number of hidden gcn units                                           | 2             |\n| Number of hidden gcn layers                                          | 1             |\n| Pseudo coordinate dimensions in GMMConv, 2 for cora and 3 for pubmed | 2             |\n| Number of kernels in GMMConv layer                                   | 3             |\n| Weight for L2 loss                                                   | 5e-4 (0.0005) |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## GAT\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value |\n| --------- | ----- |\n| Accuracy  | 0.753 |\n| Precision | 0.726 |\n| Recall    | 0.684 |\n| F-Score   | 0.697 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter                             | Value       |\n| ------------------------------------------ | ----------- |\n| Number of training epochs                  | 1000        |\n| Number of hidden attention heads           | 4           |\n| Uumber of output attention heads           | 1           |\n| Number of hidden layers                    | 1           |\n| Number of hidden units                     | 200         |\n| Use residual connection                    | False       |\n| Input feature dropout                      | 0           |\n| Attention dropout                          | 0           |\n| Learning rate                              | 1e-2 (0.01) |\n| Weight decay                               | 0           |\n| The negative slope of leaky relu           | 0.2         |\n| Indicates whether to use early stop or not | False       |\n| Skip re-evaluate the validation set        | False       |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## GCN\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value              |\n| --------- | ------------------ |\n| Accuracy  | 0.8024691358024691 |\n| Precision | 0.8550170307357392 |\n| Recall    | 0.786734693877551  |\n| F-Score   | 0.8035058227176454 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter              | Value       |\n| --------------------------- | ----------- |\n| Dropout probability         | 0           |\n| Learning rate               | 1e-2 (0.01) |\n| Number of training epochs   | 4000        |\n| Number of hidden gcn units  | 500         |\n| Number of hidden gcn layers | 1           |\n| Weight for L2 loss          | 0           |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## APPNP\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value              |\n| --------- | ------------------ |\n| Accuracy  | 0.8888888888888888 |\n| Precision | 0.9251082251082252 |\n| Recall    | 0.8943877551020407 |\n| F-Score   | 0.9049666689418242 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter              | Value         |\n| --------------------------- | ------------- |\n| Input feature dropout       | 0.25          |\n| Edge propagation dropout    | 0.5           |\n| Learning rate               | 1e-1 (0.1)    |\n| Number of training epochs   | 800           |\n| Hidden unit sizes for appnp | [64]          |\n| Number of propagation steps | 10            |\n| Teleport Probability        | 0.4           |\n| Weight for L2 loss          | 5e-4 (0.0005) |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## GIN\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value              |\n| --------- | ------------------ |\n| Accuracy  | 0.8271604938271605 |\n| Precision | 0.8205627705627706 |\n| Recall    | 0.8091836734693878 |\n| F-Score   | 0.8045525902668759 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter            | Value            |\n| ------------------------- | ---------------- |\n| Extra args                | [16, 1, 0, True] |\n| Learning rate             | 1e-2 (0.01)      |\n| Weight decay              | 5e-6 (0.000005)  |\n| Number of training epochs | 800              |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## TAGCN\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value              |\n| --------- | ------------------ |\n| Accuracy  | 0.9012345679012346 |\n| Precision | 0.9070381998953428 |\n| Recall    | 0.9102607709750566 |\n| F-Score   | 0.9041060526774812 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter            | Value                |\n| ------------------------- | -------------------- |\n| Extra args                | [16, 1, F.relu, 0.5] |\n| Learning rate             | 1e-2 (0.01)          |\n| Weight decay              | 5e-4 (0.0005)        |\n| Number of training epochs | 800                  |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## SGC\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value              |\n| --------- | ------------------ |\n| Accuracy  | 0.8271604938271605 |\n| Precision | 0.8362389490209041 |\n| Recall    | 0.8315759637188209 |\n| F-Score   | 0.8240298807695121 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter            | Value            |\n| ------------------------- | ---------------- |\n| Extra args                | [None, 1, False] |\n| Learning rate             | 1e-1 (0.1)       |\n| Weight decay              | 0                |\n| Number of training epochs | 4000             |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## AGNN\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value              |\n| --------- | ------------------ |\n| Accuracy  | 0.8765432098765432 |\n| Precision | 0.88992673992674   |\n| Recall    | 0.8888321995464853 |\n| F-Score   | 0.8850179383028748 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter            | Value                    |\n| ------------------------- | ------------------------ |\n| Extra args                | [100, 1, 1.0, True, 0.1] |\n| Learning rate             | 1e-1 (0.1)               |\n| Weight decay              | 0                        |\n| Number of training epochs | 200                      |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## ChebNet\n\n\u003ctable\u003e\n\u003ctr\u003e\u003cth\u003eStatistics\u003c/th\u003e\u003cth\u003eHyperparameters\u003c/th\u003e\u003c/tr\u003e\n\u003ctr\u003e\u003ctd\u003e\n\n| Statistic | Value              |\n| --------- | ------------------ |\n| Accuracy  | 0.9012345679012346 |\n| Precision | 0.9022735409953455 |\n| Recall    | 0.9201814058956915 |\n| F-Score   | 0.9073651359365645 |\n\n\u003c/td\u003e\u003ctd\u003e\n\n| Hyperparameter            | Value            |\n| ------------------------- | ---------------- |\n| Extra args                | [32, 1, 2, True] |\n| Learning rate             | 1e-2 (0.001)     |\n| Weight decay              | 5e-4 (0.0005)    |\n| Number of training epochs | 800              |\n\n\u003c/td\u003e\u003c/tr\u003e \u003c/table\u003e\n\n## Feature Engineering\n\n```sh\npython generate_data.py --num_classes=6\npython feature_engineering.py\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Foniani%2Fcancer_research_gnn","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Foniani%2Fcancer_research_gnn","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Foniani%2Fcancer_research_gnn/lists"}