FreePublished on GitHub

cogdl

Routes CogDL graph-learning workflows for experiments, graph data, models, training wrappers, and pipeline apps.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill

See how to install

Files included (37)

SKILL.md2.9 KB
references/package-overview.md3.2 KB
references/repo-provenance.md1.4 KB
references/repo-routing-metadata.json478 B
references/troubleshooting.md3.8 KB
scripts/check_cogdl_environment.py3.0 KB
sub-skills/experiments-and-cli/references/api-and-cli.md4.2 KB
sub-skills/experiments-and-cli/references/troubleshooting.md3.7 KB
sub-skills/experiments-and-cli/references/workflows.md3.7 KB
sub-skills/experiments-and-cli/scripts/cogdl_cli_smoke.py1.6 KB
sub-skills/experiments-and-cli/scripts/run_tiny_experiment.py4.0 KB
sub-skills/experiments-and-cli/SKILL.md2.9 KB
sub-skills/graph-data-and-datasets/references/built-in-datasets.md7.2 KB
sub-skills/graph-data-and-datasets/references/data-formats.md6.8 KB
sub-skills/graph-data-and-datasets/references/graph-and-dataset-api.md9.6 KB
sub-skills/graph-data-and-datasets/references/troubleshooting.md6.1 KB
sub-skills/graph-data-and-datasets/scripts/create_tiny_graph_dataset.py10.4 KB
sub-skills/graph-data-and-datasets/scripts/validate_graph_masks.py13.3 KB
sub-skills/graph-data-and-datasets/SKILL.md2.7 KB
sub-skills/models-layers-and-operators/references/custom-gnn.md2.4 KB
sub-skills/models-layers-and-operators/references/model-registry-and-layers.md3.1 KB
sub-skills/models-layers-and-operators/references/operator-compatibility.md1.9 KB
sub-skills/models-layers-and-operators/references/troubleshooting.md2.2 KB
sub-skills/models-layers-and-operators/scripts/custom_gnn_smoke.py1.7 KB
sub-skills/models-layers-and-operators/scripts/inspect_model_registry.py1.3 KB
sub-skills/models-layers-and-operators/SKILL.md2.7 KB
sub-skills/pipelines-and-applications/references/oagbert.md1.9 KB
sub-skills/pipelines-and-applications/references/pipeline-recipes.md2.8 KB
sub-skills/pipelines-and-applications/references/troubleshooting.md2.9 KB
sub-skills/pipelines-and-applications/scripts/pipeline_smoke.py1.7 KB
sub-skills/pipelines-and-applications/SKILL.md2.3 KB
sub-skills/training-wrappers-and-customization/references/configuration-and-best-configs.md2.2 KB
sub-skills/training-wrappers-and-customization/references/trainer-and-wrappers.md3.5 KB
sub-skills/training-wrappers-and-customization/references/troubleshooting.md2.5 KB
sub-skills/training-wrappers-and-customization/scripts/inspect_wrapper_match.py1.3 KB
sub-skills/training-wrappers-and-customization/scripts/trainer_config_template.py2.9 KB
sub-skills/training-wrappers-and-customization/SKILL.md2.6 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

CogDL

CogDL is a graph deep learning toolkit for node classification, graph classification, link prediction, graph embedding, heterogeneous graphs, traffic prediction, and application-style pipelines such as dataset inspection and OAG-BERT.

Use this root router when a user asks about CogDL itself, names one of its public APIs, or needs help choosing which part of the package to open first.

Quick start

Install PyTorch first with a CPU or CUDA wheel that matches the host.
Install CogDL in editable mode for local inspection:

pip install -e .

For a published wheel instead of a checkout install, use:

pip install cogdl

If the task needs optional package families, add them explicitly:
- ogb for OGB datasets and benchmarks
- transformers and sentencepiece for OAG-BERT
- the repo already declares optuna, gensim, grave, tabulate, numba, and ninja as runtime dependencies

Minimal import check

python -c "import cogdl; from cogdl import experiment, pipeline; print(cogdl.__version__)"
python scripts/check_cogdl_environment.py --show-registries

Use python -m pip check after install to catch incompatible dependencies.

Route map

sub-skills/experiments-and-cli/: experiment(), get_default_args, CLI flags, scripts/train.py, variants, checkpoint/log/embedding flags, and AutoML.
sub-skills/graph-data-and-datasets/: Graph, Adjacency, Dataset, NodeDataset, GraphDataset, DataLoader, masks, schemas, and tiny fixtures.
sub-skills/models-layers-and-operators/: model registry, layers, custom GNNs, and sparse/message operators.
sub-skills/training-wrappers-and-customization/: Trainer, wrappers, default wrapper matching, configs, checkpoint/resume, and logging.
sub-skills/pipelines-and-applications/: pipeline() apps, embedding generation, recommendation, and OAG-BERT.

Shared references

Read these before refreshing the skill or checking whether a checkout is current:

references/package-overview.md
references/troubleshooting.md
references/repo-provenance.md

Shared smoke script

Run scripts/check_cogdl_environment.py when you want a fast, no-download check of the installed package, registry counts, and optional CUDA/Graph smoke probes.

Scope notes

Built-in datasets may download or populate caches on first use.
OAG-BERT weights and archives are optional, cache- or network-dependent resources.
CUDA acceleration is optional in this skill tree; CPU-safe workflows are the baseline unless a sub-skill explicitly says otherwise.
For maintainer-only repository edits, CI, or release tasks, use a different workflow; this runtime skill is for research-side usage.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

More skills from this repository

3d-resnets-pytorch

Free

Routes 3D ResNets PyTorch video action-recognition workflows across training, inference, and data preparation.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill339Updated Sep 3, 2026

3ddfa

Free

Guide 3DDFA Python inference, geometry rendering, training/evaluation, and optional C++ ONNX workflows for 3D dense face alignment.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill339Updated Sep 3, 2026

3ddfa-v2

Free

Routes 3DDFA_V2 face-alignment setup, still-image demos, video tracking, and ONNX benchmarking workflows.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill339Updated Sep 3, 2026

ab3dmot

Free

Operate AB3DMOT 3D multi-object tracking workflows for KITTI and nuScenes data, tracking, evaluation, and visualization.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill339Updated Sep 3, 2026

accelerate

Free

Use Hugging Face Accelerate for PyTorch training-loop migration, distributed launch/configuration, DeepSpeed/FSDP/TPU backend setup, big-model inference/offload, checkpointing, tracking, and troubleshooting.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill339Updated Sep 3, 2026

acme

Free

Route Acme reinforcement-learning framework tasks across core loops, replay/data, JAX agents, and TensorFlow/Sonnet agents.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill339Updated Sep 3, 2026

See all skills by VectorSpaceLab

Report a problem with this skill

Install

Install with a command

npx skills add https://github.com/VectorSpaceLab/AREX-Skill --skill cogdl

Paste it into your terminal and run it (requires Node.js). You can choose which agents to install it for, such as Claude Code or Codex.

Ask your agent

Please install the following Agent Skill.
https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/cogdl

First review SKILL.md and the bundled files and confirm there is nothing suspicious (such as sending data externally or reading credentials). Then place it in ~/.claude/skills/cogdl/ for Claude Code, or ~/.agents/skills/cogdl/ for Codex.

Send this text to Claude Code or Codex as is. It checks the contents before installing.

View source on GitHub

Repository: VectorSpaceLab/AREX-Skill
Stars: 339
Forks: 26
Category: Data & AI
License: Apache-2.0
Original language: English
Last updated: Sep 3, 2026

A skill is a set of instructions and scripts for an agent. cccskills does not guarantee that skills are safe. Review the contents before you install.