giant/CLAUDE.md

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Commands

uv sync              # install dependencies (including dev extras: uv sync --extra dev)
pytest               # run tests
python scripts/train.py --data path/to/steps.parquet --mode flow   # train (flow matching)
python scripts/train.py --data path/to/steps.parquet --mode ddpm   # train (DDPM baseline)

Architecture

GIANT is a conditional generative surrogate for the Geant4 step function. It replaces the stochastic physics engine: given a pre-step particle state (conditioning), it samples a post-step outcome.

Data pipeline (giant/data/): parquet files from miniCaloSim are loaded into numpy arrays (loader.py), then log-transformed and rotated into a local coordinate frame where pre_dir = ẑ (transforms.py), before being wrapped in a PyTorch Dataset (dataset.py). Train/val split is by event_id to avoid leaking correlated steps from the same shower.

Output space (6D): step_length (log), ΔE (log), edep (log), and post_dir as a unit vector in the local frame.

Conditioning vector: PDG code (embedding), pre-step position, log(pre-energy), pre-step direction, material (embedding), layer ID, number of secondaries (Phase 1 only — see Roadmap below).

Model (giant/model/): DenoisingMLP built from ResBlocks, with a SinusoidalEmbedding for the diffusion/flow time variable and a ConditionEncoder that fuses all conditioning inputs. schedule.py provides both a CosineSchedule for DDPM and the flow matching loss utilities (Lipman et al. 2022 conditional flow matching).

Samplers (giant/sample.py): DDPM, DDIM, and flow matching (ODE integration, ~10 steps). Flow matching is the primary mode.

Validation (giant/validate.py): step-level marginal comparisons; shower-level rollout validation is planned.

Roadmap

Phase 1 (current): number of secondaries is a conditioning input — model predicts only 6D post-step kinematics.

Phase 2 (target): model must jointly predict the number of secondaries and all their properties (energy, direction, species), requiring an extended output space and likely a set-based or autoregressive generation scheme for the variable-length secondary list.