timlawrenz/gnn-ruby-code-study
GNN Ruby Code Study Systematic study of Graph Neural Network architectures for Ruby code complexity prediction and generation. Paper: Graph Neural Networks for Ruby Code Complexity Prediction and Generation: A Systematic Architecture Study Dataset 22,452 Ruby methods parsed into AST graphs with 74-dimensional node features. Split Samples File Train 19,084 dataset/train.jsonl Validation 3,368 dataset/val.jsonl Each JSONL record contains:… See the full description on the dataset page: https://huggingface.co/datasets/timlawrenz/gnn-ruby-code-study.
0101
1# Research Spec: GNN Architecture Search for Code Complexity Prediction2#3# Automated search across 5 GNN architectures and hyperparameters4# for predicting Ruby method complexity from AST structure.5#6# Launch:7# ratiocinator research specs/gnn_complexity.yaml8#9# The dataset (22K Ruby methods) is in the repo branch.10# No external data staging needed.11 12# What to research13topic: "Comparing GNN architectures (GCN, SAGE, GAT, GIN, GraphConv) for predicting Ruby code complexity from Abstract Syntax Trees. The baseline uses GraphSAGE with hidden_dim=64, 3 layers, achieving MAE 4.77 on 22K Ruby methods."14goal_metric: val_mae15maximize: false16 17# Target codebase18repo_url: https://github.com/timlawrenz/jubilant-palm-tree.git19repo_branch: experiment/ratiocinator-gnn-study20runner_script: scripts/run_complexity_arm.sh21 22# Infrastructure — models are tiny (~50K params), training is fast23hardware:24 gpu: "RTX 4090"25 num_gpus: 126 min_cpu_ram_gb: 3227 min_inet_down: 1000.028 min_cuda_version: 12.029 max_dph: 0.4030 disk_gb: 50.031 image: pytorch/pytorch:2.7.0-cuda12.8-cudnn9-runtime32 33data:34 source: none # Dataset is in the repo branch35 36deps:37 pre_install:38 - "apt-get update -qq && apt-get install -y -qq git-lfs > /dev/null 2>&1 || true"39 - "cd /workspace/experiment && git lfs install && git lfs pull"40 - "pip install torch-geometric torch-scatter torch-sparse -f https://data.pyg.org/whl/torch-2.7.0+cu128.html"41 - "pip install pandas tqdm sentence-transformers nltk scikit-learn numpy"42 requirements: requirements.txt43 exclude_from_requirements:44 - torch45 - torchvision46 - torch_geometric47 verify: "python -c \"import torch_geometric; print(f'PyG {torch_geometric.__version__}')\""48 49metrics:50 protocol: json_line51 json_prefix: "METRICS:"52 53# Budget — small models, fast training (~2-5 min per arm)54max_iterations: 355max_dollars: 15.0056train_timeout_s: 120057download_timeout_s: 60058 59# Output60paper_title: "What Graph Neural Networks Can and Cannot Learn About Code: A Systematic Empirical Study on Ruby AST Analysis"61