lars
e505e2d141
The relative path assumed a sibling checkout; point to the actual GitLab repo instead. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
100 lines
5.1 KiB
Markdown
100 lines
5.1 KiB
Markdown
# giant
|
||
|
||
**G**eant4 **I**nference via **A**utoregressive **N**eural s**T**ep surrogate — a play on *Geant4* and the step function being the computationally heaviest part of the simulation.
|
||
|
||
Proof-of-concept surrogate model for the Geant4 step function. Given a pre-step particle state, the model samples a physically plausible post-step outcome — replacing the stochastic Geant4 physics engine with a trained conditional generative model.
|
||
|
||
Training is driven entirely from parquet files of the miniCaloSim steps tree. No Geant4 runtime dependency.
|
||
|
||
## Architecture
|
||
|
||
Conditional **flow matching** model (Lipman et al. 2022): a small MLP learns a vector field mapping noise → step outcomes in ~10 ODE steps per sample. Falls back to DDPM for comparison.
|
||
|
||
**Output space (9D, diffused):**
|
||
|
||
| Index | Variable | Transform |
|
||
|-------|----------|-----------|
|
||
| 0 | `step_length` [mm] | log |
|
||
| 1 | `ΔE = pre_E − post_E` [MeV] | log |
|
||
| 2 | `edep` [MeV] | log |
|
||
| 3–5 | `post_dir` in local frame | unit vector |
|
||
| 6–8 | `travel_dir` (`post_pos − pre_pos`) in local frame | unit vector |
|
||
|
||
Both `post_dir` and `travel_dir` are expressed in the coordinate frame where `pre_dir = ẑ`, making the scattering distribution nearly azimuthally symmetric. `post_pos` itself is not a raw target — it's reconstructed at inference as `pre_pos + step_length * world_frame(travel_dir)`, so the two stay consistent by construction instead of being learned (and potentially diverging) independently.
|
||
|
||
**Conditioning:** 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).
|
||
|
||
## Roadmap
|
||
|
||
The model is developed in two phases:
|
||
|
||
**Phase 1 (current):** The number of secondaries produced in each step is passed as a conditioning input. This makes training easier because the model has direct access to multiplicity information and can focus on learning the continuous post-step kinematics.
|
||
|
||
**Phase 2 (target):** The number of secondaries is not given — the model must predict it jointly with all secondary properties (energy, direction, species) for each step. This requires extending the output space and likely an autoregressive or set-based generative approach for the variable-length secondary list.
|
||
|
||
## Data
|
||
|
||
Input: parquet files produced by [miniCaloSim](https://gitlab.etp.kit.edu/lbogner/minicalosim), or converted from a ROOT file via `scripts/steps_to_parquet.py`. Each row is one Geant4 step. Train/val split is by `event_id` (not row shuffle) to avoid leaking correlated steps from the same shower.
|
||
|
||
## Project structure
|
||
|
||
```
|
||
giant/
|
||
├── giant/
|
||
│ ├── data/
|
||
│ │ ├── loader.py # parquet → numpy arrays (incl. streaming/chunked reads)
|
||
│ │ ├── transforms.py # log transforms, local-frame rotation, normaliser
|
||
│ │ └── dataset.py # StepsDataset / StreamingStepsDataset (PyTorch)
|
||
│ ├── model/
|
||
│ │ ├── network.py # SinusoidalEmbedding, ConditionEncoder, DenoisingMLP
|
||
│ │ └── schedule.py # CosineSchedule (DDPM) and flow matching utilities
|
||
│ ├── config.py # default hyperparameters, TOML config merging, device autodetect
|
||
│ ├── pipeline.py # builds datasets/normalizers and kicks off a training run
|
||
│ ├── train.py # training loop, checkpointing, graceful shutdown
|
||
│ ├── sample.py # DDPM / DDIM / flow matching samplers
|
||
│ ├── validate.py # step-level marginal + KL-divergence validation
|
||
│ ├── analysis.py # notebook diagnostics: marginals, correlations, constraint checks
|
||
│ └── cli.py # `giant train` / `giant predict` Typer app
|
||
├── scripts/
|
||
│ ├── train.py # argparse training entry point
|
||
│ └── steps_to_parquet.py # ROOT → parquet conversion (uproot/awkward/polars)
|
||
└── tests/
|
||
```
|
||
|
||
## Setup
|
||
|
||
```bash
|
||
uv sync --extra cpu # CPU-only torch (use --extra cuda for CUDA 11.8 instead)
|
||
uv sync --extra cpu --extra dev # add dev tools (pytest, ruff, ty)
|
||
```
|
||
|
||
`cpu` and `cuda` are mutually exclusive extras selecting the torch build; plain `uv sync` installs no torch at all. See `CLAUDE.md` for details.
|
||
|
||
## Training
|
||
|
||
```bash
|
||
python scripts/train.py --data path/to/steps.parquet --mode flow
|
||
```
|
||
|
||
or via the installed CLI:
|
||
|
||
```bash
|
||
giant train path/to/steps.parquet --mode flow
|
||
giant predict path/to/steps.parquet --checkpoint checkpoints/.../best.pt
|
||
```
|
||
|
||
Both accept a TOML config file (`--config`) and CLI overrides for hyperparameters; see `--help` on either for the full option list.
|
||
|
||
## Validation
|
||
|
||
`giant.validate.validate_marginals` runs step-level marginal and KL-divergence checks during training (`--validate-every`). For deeper, notebook-driven diagnostics on a trained checkpoint — stratified marginals, correlation structure, physical constraint violations — see `giant.analysis`.
|
||
|
||
## Development
|
||
|
||
```bash
|
||
uv run pytest # run tests
|
||
uv run ruff check . # lint
|
||
uv run ruff format . # format
|
||
uv run ty check . # type check
|
||
```
|