dct-toolkit 0.4.0__tar.gz

dct_toolkit-0.4.0/.gitignore

@@ -0,0 +1,58 @@
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Testing
+.pytest_cache/
+.coverage
+htmlcov/
+.tox/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Jupyter
+.ipynb_checkpoints/
+*.ipynb
+
+# Environment
+.env
+.venv
+env/
+venv/
+ENV/
+
+# Documentation builds
+docs/_build/
+site/
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json

dct_toolkit-0.4.0/AGENTS.md

@@ -0,0 +1,417 @@
+# AGENTS.md - DCT Toolkit Agent Guidelines
+
+**Project**: dct-toolkit  
+**Version**: 0.4.0  
+**Status**: Private Repository in Publication Preparation  
+**Last Updated**: 2026-04-10
+
+---
+
+## Critical Rule: Standalone Project
+
+⚠️ **THIS IS A COMPLETELY STANDALONE PROJECT**
+
+- ✅ ONLY import from within `dct_toolkit/dct_toolkit/` directory
+- ✅ External dependencies: `numpy`, `scipy` ONLY
+- ❌ NEVER import from parent directories (`../src/`, `../dct_stats/`, legacy code)
+- ❌ NEVER import from any code outside this repository
+- ❌ NEVER assume other projects are available
+
+**Verification Command**:
+```bash
+grep -r "^from \.\./" dct_toolkit/
+grep -r "^import .*\.\." dct_toolkit/
+```
+Should return NO matches.
+
+---
+
+## Environment Setup
+
+### Conda Environment
+```bash
+conda activate myenv
+python --version  # Should be 3.8+
+```
+
+### PYTHONPATH
+When running tests or examples from project root:
+```bash
+export PYTHONPATH=$PYTHONPATH:$(pwd)/dct_toolkit
+```
+
+### Isolated Conda Environment for Testing/Building (Recommended)
+To avoid dependency conflicts when building or testing the package, use an isolated conda environment:
+
+```bash
+# Create isolated environment
+conda create -n dct-toolkit-dev python=3.11 -y
+conda activate dct-toolkit-dev
+
+# Install package in development mode
+pip install -e .
+
+# Run tests
+python -m pytest tests -q
+
+# Build package artifacts (for PyPI)
+conda install -c conda-forge build twine -y
+python -m build
+python -m twine check dist/*
+
+# Clean up when done
+conda deactivate
+```
+
+**Why isolated?** Building tools (`build`, `twine`) and their dependencies can conflict with other packages in your base environment (e.g., Sphinx, Streamlit). Always use an isolated environment for package operations.
+
+---
+
+## Code Style - Scientific Repository Standards
+
+### Docstrings (REQUIRED)
+Use NumPy style:
+```python
+def function_name(param1: type, param2: type) -> return_type:
+    """
+    Short description.
+    
+    Longer description if needed.
+    
+    Parameters
+    ----------
+    param1 : type
+        Description
+    param2 : type
+        Description
+        
+    Returns
+    -------
+    return_type
+        Description
+        
+    Examples
+    --------
+    >>> result = function_name(1.0, 2.0)
+    """
+```
+
+### Type Hints (REQUIRED)
+All function parameters and returns must be typed:
+```python
+def smooth_data(data: np.ndarray, width: float) -> np.ndarray:
+    ...
+```
+
+### Naming Conventions
+- **Functions/Variables**: `snake_case` (e.g., `get_transfer_function`)
+- **Classes**: `CamelCase` (e.g., `DCTSmoother`)
+- **Constants**: `UPPER_SNAKE_CASE` (e.g., `DEFAULT_KERNEL`)
+- **Private**: `_leading_underscore` (e.g., `_helper_func`)
+
+### Formatting
+- **Indentation**: 4 spaces (NO tabs)
+- **Line Length**: 100 characters (soft limit)
+- **Blank Lines**: 
+  - 2 lines between top-level functions/classes
+  - 1 line between methods
+- **Operators**: Spaces around operators (`x = y + 1`)
+
+### Import Organization
+```python
+# 1. Standard library
+import os
+import sys
+from typing import Tuple, Optional
+
+# 2. Third-party (numpy, scipy)
+import numpy as np
+import scipy.fft
+
+# 3. Local modules (relative imports)
+from .core import get_dct_transfer_function
+from .stats import dct_mean
+```
+
+---
+
+## Project Structure
+
+```
+dct_toolkit/                    # PROJECT ROOT
+│
+├── dct_toolkit/               # INSTALLABLE PACKAGE ⚠️ CODE GOES HERE
+│   ├── __init__.py           # Public API exports
+│   ├── core.py               # Transfer functions & 1D primitives
+│   ├── cartesian.py          # 2D Cartesian separable smoothing
+│   ├── polar.py              # 2D Polar smoothing (adaptive kernels)
+│   ├── stats.py              # Statistical ops (Normalized Convolution)
+│   └── gap_filling.py         # Gap-filling methods (deferred for public stats-first release)
+│
+├── tests/                     # TEST SUITE ⚠️ TESTS GO HERE
+│   ├── test_core.py
+│   ├── test_cartesian.py
+│   ├── test_polar.py
+│   ├── test_stats.py
+│   └── test_gap_filling.py
+│
+├── examples/                  # USAGE EXAMPLES
+│   ├── basic_polar.py
+│   └── comprehensive_demo.py
+│
+├── exp_v3/                    # EXPERIMENTS (reports + code)
+│   ├── figures/               # Generated figures (PNG/PDF)
+│   ├── test_width_impact.py
+│   ├── plot_gap_filling_results.py
+│   ├── GAP_FILLING_GUIDE.md
+│   ├── TEST_REPORT_GAP_FILLING.md
+│   ├── TEST_REPORT_CORE.md
+│   └── gap_filling_results.csv
+│
+├── docs/                      # DOCUMENTATION
+│   ├── API_REFERENCE.md
+│   ├── MATHEMATICAL_BASIS.md
+│   └── GAP_FILLING_BASIS.md
+│
+├── README.md                  # Main documentation
+├── CHANGELOG.md               # Version history
+└── AGENTS.md                  # This file
+```
+
+---
+
+## Development Workflow
+
+### Before Making Changes
+
+1. **Activate Environment**
+   ```bash
+   conda activate myenv
+   ```
+
+2. **Verify Baseline**
+   ```bash
+   export PYTHONPATH=$PYTHONPATH:$(pwd)/dct_toolkit
+   python -m pytest tests/ -v
+   ```
+   All tests in `tests/` must pass before you start.
+
+3. **Review Documentation**
+   - Check `docs/MATHEMATICAL_BASIS.md` for theory
+   - Check `docs/GAP_FILLING_BASIS.md` for gap filling theory
+   - Check `docs/API_REFERENCE.md` for existing API
+
+### Making Changes
+
+1. **Edit Code**
+   - Modify appropriate module in `dct_toolkit/dct_toolkit/`
+   - Follow code style guidelines
+   - Add docstrings and type hints
+
+2. **Add Tests**
+   - Add/update tests in `dct_toolkit/tests/`
+   - Tests should validate correctness with synthetic data
+   - Include edge cases (empty arrays, all NaN, etc.)
+
+3. **Update Documentation**
+   - Update `API_REFERENCE.md` if API changes
+   - Update `CHANGELOG.md` with changes
+   - Update this file if workflow changes
+
+### Testing (MANDATORY)
+
+**Run Tests**:
+```bash
+export PYTHONPATH=$PYTHONPATH:$(pwd)/dct_toolkit
+python -m pytest tests/ -v
+```
+
+**Expected Output**: all tests passing
+
+**If Tests Fail**:
+- Fix the code, not the tests
+- Tests represent ground truth requirements
+- Never commit with failing tests
+
+### Pre-Completion Checklist
+
+Before declaring task complete:
+
+- [ ] All tests pass (`pytest tests/ -v`)
+- [ ] No imports from outside `dct_toolkit/`
+- [ ] All public functions have NumPy-style docstrings
+- [ ] All functions have type hints
+- [ ] Code follows naming conventions
+- [ ] No tabs, 4-space indentation
+- [ ] Line length ≤ 100 characters
+- [ ] CHANGELOG.md updated
+- [ ] Documentation updated if API changed
+
+---
+
+## Current Progress & Roadmap
+
+### ✅ Completed (v0.4.0)
+- [x] Core primitives (`core.py`)
+  - Transfer functions (boxcar, boxcar_discrete, gaussian)
+  - 1D DCT convolution
+- [x] Cartesian smoothing (`cartesian.py`)
+  - Separable N-D smoothing
+- [x] Polar smoothing (`polar.py`)
+  - Adaptive azimuth kernels
+  - Boundary conditions (reflective/periodic)
+- [x] Statistical operations (`stats.py`)
+  - Normalized Convolution
+  - dct_count, dct_mean, dct_prefill, dct_variance, dct_std
+- [x] Gap-filling module (`gap_filling.py`)
+  - iterative diffusion fill (`iterative_gap_fill`)
+  - DCT-PLS inpainting (`dct_inpaint`)
+- [x] Test suite (`tests/`) passing
+- [x] Documentation suite
+
+### 🚧 In Progress: Publication Readiness (stats-first)
+- [x] Fix immediate correctness issues in core statistics path
+- [x] Create a dedicated publication-prep branch
+- [ ] Update `AGENTS.md` / docs to reflect publication plan and current state
+- [ ] Build clean public-facing scope focused on convolution/statistics
+
+### 📋 Before Public Release
+- [x] Add packaging metadata (`pyproject.toml`)
+- [x] Add `LICENSE` and align README badges/claims
+- [ ] Define public API surface (exclude gap filling for initial public release)
+- [ ] Remove or relocate experimental assets from public-facing surface
+- [x] Add CI for tests and minimum quality checks
+
+### 🔭 Future (post-public v1)
+- [ ] 3D support (volumetric data)
+- [ ] Additional kernels (Savitzky-Golay, Hanning)
+- [ ] Performance optimizations (numba?)
+- [ ] Jupyter notebook examples
+- [ ] conda-forge distribution
+- [ ] Advanced gap filling (constrained optimization)
+- [ ] GPU acceleration (CuPy)
+
+---
+
+## Key Technical Details
+
+### Transfer Functions
+- **Boxcar (Analytical)**: `H[k] = sin(W·θ)/(W·sin(θ))`, θ = πk/(2n)
+- **Boxcar (Discrete)**: Sum of cosines for discrete kernels
+- **Gaussian**: `H[k] = exp(-0.5·(ω·σ)²)`, σ = width/√12
+
+### Normalized Convolution
+Formula for robust mean with gaps:
+```
+μ = smooth(data * mask) / (smooth(mask) + ε)
+```
+
+### Polar Coordinates
+- **Data Layout**: (n_azimuth, n_range)
+- **Adaptive Width**: w_az(r) = width / (r · dθ)
+- **Boundary Conditions**:
+  - Reflective: DCT-II (even symmetry)
+  - Periodic: Real FFT (circular convolution)
+
+### Testing Philosophy
+- Synthetic data with known ground truth
+- Verify theoretical properties (DC preservation, linearity)
+- Edge case coverage (all NaN, single point, etc.)
+
+---
+
+## Common Pitfalls & Solutions
+
+### 1. Import Errors
+**Problem**: `ModuleNotFoundError: No module named 'dct_toolkit'`
+**Solution**: 
+```bash
+export PYTHONPATH=$PYTHONPATH:$(pwd)/dct_toolkit
+```
+
+### 2. Path Confusion
+**Problem**: Tests run from wrong directory
+**Solution**: Always run from project root (`dct_toolkit/`)
+
+### 3. Polar Boundary Issues
+**Problem**: Discontinuity at 360°/0°
+**Solution**: Use `az_boundary='periodic'` for azimuth
+
+### 4. Test Failures
+**Problem**: Tests fail after changes
+**Solution**: Fix the code, don't modify tests to pass
+
+### 5. Legacy Code Contamination
+**Problem**: Accidentally importing from `../src/`
+**Solution**: Check with grep command in Critical Rule section
+
+---
+
+## Repository Management
+
+### Private Repository (Current)
+- Hosted on: GitHub (private)
+- Access: Core team only
+- Branching: feature branches → main
+- Manual testing required until CI is added
+
+### Planned Public Release Scope
+- Initial public release is **convolution/statistics focused**
+- Keep DCT + periodic FFT boundary support in smoothing/statistical APIs
+- Defer gap-filling methods from initial public API surface
+- Add packaging metadata + license before publication
+
+### Publication Assets (Current Branch)
+- `pyproject.toml` for pip builds and installation
+- `LICENSE` (MIT)
+- `.github/workflows/tests.yml` (pytest on push/PR)
+- `docs/PUBLICATION_GUIDE.md` with pip + conda-forge steps
+- `conda.recipe/meta.yaml` starter recipe for staged-recipes submission
+
+---
+
+## Emergency Contacts & References
+
+### Documentation
+- **Math Theory**: `docs/MATHEMATICAL_BASIS.md`
+- **Gap Filling Basis**: `docs/GAP_FILLING_BASIS.md`
+- **API Details**: `docs/API_REFERENCE.md`
+- **Core Test Report**: `exp_v3/TEST_REPORT_CORE.md`
+- **Gap Filling Report**: `exp_v3/TEST_REPORT_GAP_FILLING.md`
+
+### Code Examples
+- **Polar Smoothing**: `examples/basic_polar.py`
+- **Full Demo**: `examples/comprehensive_demo.py`
+- **Gap Filling**: `exp_v3/test_width_impact.py`
+
+### Key Papers & References
+- Garcia (2010): DCT-PLS (contrast/background)
+- Knutsson et al.: Normalized Convolution
+- Ahmed et al.: DCT fundamentals
+- Lecture 17 Notes: Objective Mapping / Kriging (for context)
+
+---
+
+## Quick Reference Card
+
+```bash
+# Setup
+conda activate myenv
+export PYTHONPATH=$PYTHONPATH:$(pwd)/dct_toolkit
+
+# Test
+python -m pytest tests/ -v
+
+# Run example
+python examples/basic_polar.py
+
+# Benchmark
+python exp_v3/test_width_impact.py
+
+# Check imports (should be empty)
+grep -r "^from \.\./" dct_toolkit/dct_toolkit/
+```
+
+---
+
+**Remember**: This is a standalone scientific package. Quality, correctness, and clarity are more important than speed. When in doubt, refer to the test suite - it defines the expected behavior.

dct_toolkit-0.4.0/CHANGELOG.md

@@ -0,0 +1,139 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+### Added
+- **Packaging + distribution groundwork**:
+    - Added `pyproject.toml` (PEP 517/518 via hatchling) for pip/wheel builds.
+    - Added `LICENSE` (MIT).
+    - Added GitHub Actions CI workflow `.github/workflows/tests.yml`.
+    - Added `docs/PUBLICATION_GUIDE.md` with PyPI + conda-forge release steps.
+    - Added `conda.recipe/meta.yaml` starter recipe for conda-forge submission prep.
+- **Statistics API**:
+    - Added `dct_prefill` (iterative normalized-convolution prefill) and exported it
+      in the top-level API.
+    - Added regression tests for `dct_prefill` behavior and parameter validation.
+
+### Changed
+- `stats.py`: improved dtype handling for integer inputs in `dct_mean`/`dct_variance`
+  with floating-point outputs and explicit mask-shape validation.
+- `README.md` and `docs/API_REFERENCE.md`: refocused public surface on
+  convolution/statistics and documented `dct_prefill`.
+- `examples/comprehensive_demo.py`: removed stale experimental imports and aligned
+  with stats-first public API.
+
+## [0.4.0] - 2026-02-28
+
+### Added
+- **DCT Spectral Inpainting (`dct_inpaint`)**:
+    - New gap-filling method based on DCT-domain Penalised Least Squares (Garcia, 2010).
+    - Minimises ‖W·(y − û)‖² + λ·‖Δᵖ û‖² with automatic width-to-lambda mapping.
+    - Default order p=2 (bi-harmonic) preserves curvature across gaps — equivalent to
+      thin-plate spline interpolation.
+    - Supports both Cartesian (DCT-II reflective BC) and Polar (RFFT periodic azimuth +
+      DCT reflective range) coordinates.
+    - **Adaptive polar eigenvalues**: When `az_res_deg` is provided, azimuth penalty
+      scales with range (`1/(r·dθ)^(2p)`), mirroring `smooth_polar`'s adaptive kernels.
+    - Initialises with fast linear interpolation baseline, then refines iteratively.
+    - 2.9x more accurate than `iterative_gap_fill` and 34% more accurate than
+      `scipy.interpolate.griddata` on the standard polar benchmark (wrapping hole).
+    - **Input validation**: Rejects `width <= 0`, `order < 1`, invalid `coordinates`
+      and `az_boundary` values with clear error messages.
+- **Helper functions**:
+    - `_width_to_lambda`: Derive Tikhonov lambda from smoothing width.
+    - `_eigenvalues_dct` / `_eigenvalues_dft`: Compute Laplacian eigenvalues for
+      reflective / periodic boundary conditions.
+    - `_compute_eigenvalues_2d`: Build 2-D eigenvalue tensor (isotropic or
+      polar-adaptive with range-dependent azimuth scaling).
+    - `_forward_transform` / `_inverse_transform`: Unified spectral transforms
+      supporting mixed BC (periodic azimuth + reflective range).
+- **Test suite** (`tests/test_gap_filling.py`):
+    - 42 tests covering helpers, transform round-trips, convergence, 1D/2D Cartesian,
+      polar (periodic + reflective + adaptive eigenvalues), curvature preservation,
+      input validation, edge cases, and v3 comparison.
+- **Benchmark** (`exp_v4/test_inpaint_vs_v3.py`):
+    - Reproduces exp_v3 polar benchmark with dct_inpaint added.
+- **Noisy benchmark** (`exp_v4/test_noisy_inpaint.py`):
+    - Evaluates noise robustness across SNR levels 3–100.
+- **Figures** (`exp_v4/figures/`):
+    - 8 publication-quality figures (PNG + PDF): spatial comparison (v3 vs v4 vs
+      griddata), error maps, width impact, convergence, method bar chart,
+      noisy spatial comparison, SNR sweep, and noisy width impact.
+- **Documentation**:
+    - Updated `docs/API_REFERENCE.md` with full `dct_inpaint` entry.
+    - Updated `docs/GAP_FILLING_BASIS.md` with DCT-PLS theory, eigenvalue
+      diagonalisation, width-to-lambda bridge, and benchmark results.
+    - Updated `docs/MATHEMATICAL_BASIS.md` with cross-reference to inpainting theory.
+    - Updated `README.md` with gap filling features and usage examples.
+    - Updated `exp_v4/PLAN.md` with refined mathematical formulation.
+
+### Changed
+- `__init__.py`: Exports `dct_inpaint`; version bumped to `0.4.0`.
+- `gap_filling.py`: Module docstring updated to document both algorithms.
+  `_compute_eigenvalues_2d` now accepts optional `az_res_deg` for polar-adaptive
+  eigenvalues.
+
+## [0.2.1] - 2026-02-09
+
+### Fixed
+- **Gap Filling Initialization**:
+    - Improved `init='linear'` to prioritize azimuth interpolation without extrapolation (using range interpolation for edges), preventing artifacts on periodic boundaries.
+- **Benchmark Consistency**:
+    - Unified `exp_v3/test_width_impact.py` and `exp_v3/plot_gap_filling_results.py` to use identical test scenarios (Index Space hole).
+    - Fixed MAE discrepancy in figures vs. report.
+
+## [0.2.0] - 2026-02-09
+
+### Added
+- **Gap Filling Module (`gap_filling.py`)**:
+    - `iterative_gap_fill`: Constructive gap filling via iterative DCT-based diffusion.
+    - Linear interpolation initialization (axis-wise `np.interp`, default) — preserves spatial gradients across holes from the first iterate.
+    - Three initialization modes: `'linear'` (default), `'multiscale'` (coarse-to-fine DCT cascade), `'dct'` (single-pass Normalized Convolution).
+    - `smooth_output` option for combined gap-fill + noise reduction.
+    - Convergence via relative L2 norm change tolerance.
+- **Comprehensive Gap Filling Test Suite** (`tests/test_gap_filling_comprehensive.py`):
+    - 13 tests covering accuracy, robustness, speed, convergence, init modes, backward compatibility, and edge cases.
+- **Benchmark Suite** (`exp_v3/test_width_impact.py`):
+    - 84-evaluation benchmark (3 datasets x 7 gap scenarios x 4 methods) with auto-generated CSV results.
+- **Figures** (`exp_v3/figures/`):
+    - 4 publication-quality figures (PNG + PDF): spatial comparison, width impact, iteration convergence, uncertainty maps.
+- **Test Report** (`exp_v3/TEST_REPORT_GAP_FILLING.md`):
+    - Auto-generated benchmark report; all 5 evaluation criteria PASS.
+- **Algorithm Documentation** (`exp_v3/GAP_FILLING_GUIDE.md`):
+    - Algorithm description, tuning guidance, and key findings from diagnostic analysis.
+
+### Changed
+- `__init__.py`: Exports `iterative_gap_fill`; version bumped to `0.2.0`.
+- Gap filling promoted from `experimental/` to core module `dct_toolkit/gap_filling.py`.
+
+### Deprecated
+- `multiscale` parameter in `iterative_gap_fill` — use `init='multiscale'` or `init='dct'` instead. Emits `FutureWarning`.
+
+## [0.1.0] - 2026-02-08
+
+### Added
+- **Core Package Structure**: Created `dct_toolkit` as a standalone python package.
+- **Core Primitives (`core.py`)**:
+    - Analytical and Discrete transfer functions for 'boxcar', 'boxcar_discrete', 'gaussian'.
+    - `dct_convolve_1d` primitive using DCT-II/Ortho.
+- **2D Cartesian Smoothing (`cartesian.py`)**:
+    - Separable N-D smoothing implementation.
+- **2D Polar Smoothing (`polar.py`)**:
+    - Support for (Azimuth, Range) data layout.
+    - Adaptive azimuth kernel width (maintains constant physical width).
+    - Boundary conditions: 'reflective' (DCT) and 'periodic' (Real FFT) for azimuth.
+- **Statistical Operations (`stats.py`)**:
+    - `dct_count`: Effective sample size calculation.
+    - `dct_mean`: Robust local mean using Normalized Convolution (handles NaNs).
+    - `dct_variance` & `dct_std`: Robust second-moment statistics.
+- **Experimental Gap Filling (`experimental/`)**:
+    - Iterative Constructive Gap Filler based on `dct_mean`.
+    - Benchmark script showing >100x accuracy improvement over linear interpolation for smooth fields.
+- **Validation**:
+    - Comprehensive test suite in `tests/`.
+    - Example script `examples/basic_polar.py` demonstrating boundary conditions.

dct_toolkit-0.4.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Jairo M. Valdivia
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.