hypercube-lcn 1.0.0__tar.gz

hypercube_lcn-1.0.0/CMakeLists.txt

@@ -0,0 +1,76 @@
+cmake_minimum_required(VERSION 3.20)
+project(HypercubeLCNPython LANGUAGES CXX)
+
+set(CMAKE_CXX_STANDARD 23)
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
+set(CMAKE_POSITION_INDEPENDENT_CODE ON)
+
+# ── pybind11 ──
+find_package(pybind11 CONFIG REQUIRED)
+
+# ── Optimization flags (match main project / sibling python builds) ──
+# HYPERCUBE_ARCH controls -march. Defaults to "native" for local dev builds.
+# cibuildwheel overrides to "x86-64-v2" (x86_64) or "none" (ARM, MSVC).
+set(HYPERCUBE_ARCH "native" CACHE STRING "Target architecture for -march (native, x86-64-v2, none)")
+
+if(MSVC)
+    add_compile_options(/O2 /fp:fast)
+else()
+    add_compile_options(-O3 -ffast-math)
+    if(NOT HYPERCUBE_ARCH STREQUAL "none")
+        if(HYPERCUBE_ARCH STREQUAL "native")
+            add_compile_options(-march=native -mtune=native)
+        else()
+            add_compile_options(-march=${HYPERCUBE_ARCH} -mtune=generic)
+        endif()
+    endif()
+    add_compile_options(-Wall -Wextra -Wno-unknown-pragmas)
+endif()
+
+# ── Core sources compiled directly into the module (PIC required) ──
+set(CORE_DIR "${CMAKE_CURRENT_SOURCE_DIR}/..")
+set(CORE_SOURCES
+    ${CORE_DIR}/Core.cpp
+    ${CORE_DIR}/Training.cpp
+)
+
+# ── Package version (single source: hypercube_lcn/_version.py) ──
+set(_HLCN_VERSION_FILE "${CMAKE_CURRENT_SOURCE_DIR}/hypercube_lcn/_version.py")
+set(HYPERCUBE_LCN_VERSION "")
+file(STRINGS "${_HLCN_VERSION_FILE}" _HLCN_VERSION_LINES)
+foreach(_line IN LISTS _HLCN_VERSION_LINES)
+    # Strip CR (Windows) and match: __version__ = "x.y.z"
+    string(REPLACE "\r" "" _line "${_line}")
+    if(_line MATCHES "^__version__[ \t]*=[ \t]*\"([^\"]+)\"")
+        set(HYPERCUBE_LCN_VERSION "${CMAKE_MATCH_1}")
+        break()
+    endif()
+endforeach()
+if(HYPERCUBE_LCN_VERSION STREQUAL "")
+    message(FATAL_ERROR
+        "Could not parse __version__ from hypercube_lcn/_version.py")
+endif()
+message(STATUS "hypercube_lcn version: ${HYPERCUBE_LCN_VERSION}")
+
+# ── Build the Python extension module ──
+pybind11_add_module(_core bindings.cpp ${CORE_SOURCES})
+target_include_directories(_core PRIVATE ${CORE_DIR})
+target_compile_definitions(_core PRIVATE
+    "HYPERCUBE_LCN_VERSION=\"${HYPERCUBE_LCN_VERSION}\"")
+
+# ── Linking ──
+if(MINGW)
+    # Static libgcc/libstdc++ keep the .pyd free of those DLLs. Do NOT static-link
+    # winpthread: mingw-w64 15.2.0's libwinpthread.a references __intrinsic_setjmpex
+    # and fails at link (same constraint as the sibling python builds).
+    # Posix-model MinGW still needs libwinpthread-1.dll at runtime — ship the DLL
+    # from THIS toolchain (stale copies miss symbols like nanosleep64).
+    target_link_options(_core PRIVATE -static-libgcc -static-libstdc++)
+    find_file(WINPTHREAD_DLL libwinpthread-1.dll PATHS ENV PATH NO_DEFAULT_PATH)
+    if(WINPTHREAD_DLL)
+        install(FILES ${WINPTHREAD_DLL} DESTINATION hypercube_lcn)
+    endif()
+endif()
+
+# ── Install into the hypercube_lcn package directory ──
+install(TARGETS _core DESTINATION hypercube_lcn)

hypercube_lcn-1.0.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 David Charles Liptak
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

hypercube_lcn-1.0.0/PKG-INFO

@@ -0,0 +1,329 @@
+Metadata-Version: 2.4
+Name: hypercube-lcn
+Version: 1.0.0
+Summary: Python bindings for HypercubeLCN: a locally connected network on a Boolean hypercube, every weight trained
+License-Expression: Apache-2.0
+Classifier: Development Status :: 5 - Production/Stable
+Classifier: Intended Audience :: Science/Research
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Programming Language :: C++
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Operating System :: Microsoft :: Windows
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Operating System :: MacOS
+Project-URL: Homepage, https://github.com/dliptak001/HypercubeLCN
+Project-URL: Documentation, https://github.com/dliptak001/HypercubeLCN/blob/main/docs/Python_SDK.md
+Project-URL: Repository, https://github.com/dliptak001/HypercubeLCN
+Requires-Python: >=3.10
+Requires-Dist: numpy>=1.21
+Provides-Extra: test
+Requires-Dist: pytest>=7.0; extra == "test"
+Description-Content-Type: text/markdown
+
+# Hypercube LCN
+
+This package is the **Python** surface for HypercubeLCN
+(`import hypercube_lcn`).
+Full API reference: **[docs/Python_SDK.md](https://github.com/dliptak001/HypercubeLCN/blob/main/docs/Python_SDK.md)**.
+C++ integration guide: **[docs/CPP_SDK.md](https://github.com/dliptak001/HypercubeLCN/blob/main/docs/CPP_SDK.md)**.
+Project home: **[github.com/dliptak001/HypercubeLCN](https://github.com/dliptak001/HypercubeLCN)**.
+
+HypercubeLCN is a **Locally Connected Network** on a Boolean
+hypercube: a deep feedforward network whose connectivity is the
+cube's own edges and whose weights are trained. It is built from two
+core classes.
+
+The **Core** class is the network. It owns the weights and runs the
+forward pass: one field in, one field out, with a stack of
+intermediate fields written on the same cube in between.
+
+The **Training** class walks that same pass in reverse. It
+accumulates gradients through every depth and steps the weights with
+Adam.
+
+That is the whole architecture. There is no preprocessor, no
+reservoir, no separate readout — the cube is the model. In Python the
+two are wrapped by a single class, **`hypercube_lcn.LCN`**.
+
+This is the opposite bet from the sibling projects. HypercubeEtalon,
+HypercubeWTF, and HypercubeCascade all put a *frozen random*
+hypercube stage in front of a small trained readout, and the point of
+those experiments is how far fixed dynamics can carry a thin
+classifier. This project asks instead: how much better does the
+hypercube do when every weight in it is trained?
+
+---
+
+<p align="center">
+  <strong>HypercubeAI ecosystem</strong><br/>
+</p>
+
+<p align="center">
+  <a href="https://github.com/dliptak001/HypercubeESN"><strong>HypercubeESN</strong></a>
+  &nbsp;·&nbsp;
+  <a href="https://github.com/dliptak001/HypercubeCNN"><strong>HypercubeCNN</strong></a>
+  &nbsp;·&nbsp;
+  <a href="https://github.com/dliptak001/HypercubeHopfield"><strong>HypercubeHopfield</strong></a>
+  &nbsp;·&nbsp;
+  <a href="https://github.com/dliptak001/HypercubeWTF"><strong>HypercubeWTF</strong></a>
+  &nbsp;·&nbsp;
+  <a href="https://github.com/dliptak001/HypercubeEtalon"><strong>HypercubeEtalon</strong></a>
+  &nbsp;·&nbsp;
+  <a href="https://github.com/dliptak001/HypercubeCascade"><strong>HypercubeCascade</strong></a>
+  &nbsp;·&nbsp;
+  <a href="https://github.com/dliptak001/HypercubeLCN"><strong>HypercubeLCN</strong></a>
+</p>
+
+HypercubeLCN is an experiment in the **HypercubeAI** project — our
+quest to systematically re-implement classical neural architectures
+on a Boolean hypercube topology instead of Euclidean grids or random
+graphs. The central thesis is "topology-native intelligence": the
+hypercube's algebraic structure (vertex-transitive symmetry, Hamming
+geometry, bitwise addressing) can serve as a first-class
+computational substrate.
+
+- **A topology you don’t store** — the graph is specified: connectivity is
+  implicit in the vertex indices; with a seed and a few config scalars the
+  whole reservoir reconstructs mathematically.
+- **Perfect homogeneity** — every vertex has the same degree and the same local
+  world, so local dynamics mean the same thing everywhere — no structural
+  favorites baked in by a random graph.
+- **Cheap navigation** — each neighbor is a few bit operations on the vertex
+  index, not a pointer chase through a stored edge list, so walks stay
+  arithmetic and cache-friendly.
+- **Nothing leaves the cube** — input, every intermediate field, and output
+  all live on the same vertices. Depth is the only direction anything travels.
+
+Each product in the family is a different architecture on that same
+foundation.
+
+---
+
+## The LCN
+
+A **locally connected network** is wired like a convolutional layer
+— each unit reads only a small neighborhood — but where a CNN slides
+one shared kernel across every position, an LCN lets every position
+train its own private weights.
+
+Here the hypercube supplies the neighborhoods. A vertex's neighbors
+are the indices one bit-flip away, and at every depth each vertex
+gathers from all of them at once. Each vertex owns a private weight
+table at every depth — one weight for each neighbor and each field
+that neighbor shows it — and shares nothing with any other vertex.
+The one wrinkle is a lookback window: when a vertex reads from a
+neighbor, it sees not just that neighbor's newest field but the last
+few written (a configurable width, two to six). This acts as a short
+skip connection and keeps the input visible to the early depths.
+
+Training runs the forward pass's loops in reverse, and no depth is
+spared: the gradient reaches every weight at every depth. For the full story,
+[docs/forward.md](https://github.com/dliptak001/HypercubeLCN/blob/main/docs/forward.md) and
+[docs/training.md](https://github.com/dliptak001/HypercubeLCN/blob/main/docs/training.md)
+walk through the code loop by loop, with dim-4 examples small enough
+to check by hand.
+
+---
+
+## Raman baseline extraction (a vibrational spectroscopy application)
+
+The benchmark is the one the sibling projects established: recover
+the slow fluorescence background under sharp molecular peaks without
+lifting the baseline into the bands or cutting trenches beneath them.
+The dataset is 10,000 synthetic LiCoO₂ (lithium cobalt oxide)
+training spectra and 2,000 held-out validation spectra, scored as
+RMSE in raw counts.
+
+On this task the frozen-stage siblings — Etalon, WTF, and Cascade,
+each feeding the same one-layer, one-channel readout — all landed on
+one floor: 4.76 to 4.82 validation. The LCN, with a dim-11 cube
+matched to the 2048-bin spectrum and every weight trained, scores
+**1.98 training / 2.03 validation**.
+
+![Held-out validation extract, spectra 351 through 354](https://raw.githubusercontent.com/dliptak001/HypercubeLCN/main/examples/RamanBaseline/extracted_baselines.png)
+
+Grey is the raw spectrum, red the true baseline, blue the extract. At
+two counts of RMSE the residual is at the scale of the label's own
+noise, and the red trace all but disappears under the blue. A second
+experiment widens the cube to dim 12 so half the vertices serve as
+free hidden units, and scores **1.64 / 1.69**. The write-ups are
+[examples/RamanBaseline/](https://github.com/dliptak001/HypercubeLCN/blob/main/examples/RamanBaseline/README.md) and
+[examples/RamanBaselineNarrowIO/](https://github.com/dliptak001/HypercubeLCN/blob/main/examples/RamanBaselineNarrowIO/README.md).
+
+---
+
+## Installation
+
+**Preferred:** install a pre-built wheel from PyPI (no compiler).
+
+```bash
+pip install hypercube-lcn
+```
+
+```python
+import hypercube_lcn as hl
+print(hl.__version__)
+```
+
+Package name on PyPI: **`hypercube-lcn`**. Import name:
+**`hypercube_lcn`**. Main type: **`hl.LCN`**.
+
+Wheels target Python 3.10–3.14 on common Windows, Linux, and macOS machines.
+Runtime dependency: NumPy only.
+
+### From source (full repository)
+
+To compile the extension yourself, clone this **entire** repository (not a
+minimal source-only download of the `python/` folder alone — the C++ core
+lives next to `python/`). You need Python 3.10+, a C++23 compiler, and
+CMake ≥ 3.20.
+
+```bash
+git clone https://github.com/dliptak001/HypercubeLCN.git
+cd HypercubeLCN/python
+pip install .
+```
+
+On Windows with CLion's MinGW, put that compiler's `bin` folder (and Ninja) on
+your `PATH`, then:
+
+```bash
+pip install . --no-build-isolation --force-reinstall --no-deps
+```
+
+(Exact CLion paths change with the version.)
+
+---
+
+## Quick start
+
+You bring each sample as a length-**N** float array (N = 2<sup>dim</sup>). How
+you get there — pad an image, reshape a spectrum, invent a layout — is up to
+you. This package does not pack 784 pixels or 2048 bins for you.
+
+Shapes that matter:
+
+| Array | Shape | Notes |
+|-------|-------|-------|
+| `fields` | `(count, N)` | one length-N field per row |
+| `targets` | `(count, width)` | width ≤ N; width < N masks the loss to vertices 0..width-1 |
+
+```python
+import numpy as np
+import hypercube_lcn as hl
+
+dim = 6
+N = 2**dim
+rng = np.random.default_rng(0)
+fields = rng.standard_normal((256, N)).astype(np.float32)
+targets = 0.5 * (fields + fields[:, np.arange(N) ^ 1])  # a local map
+
+net = hl.LCN(dim=dim, gather_span=3, tanh_last=False,
+             lr=1e-2, lr_min_frac=0.05, restore_best=True)
+net.fit(fields, targets, epochs=40, batch_size=16, verbose=True)
+
+prediction = net.forward(fields[0])   # (N,) float32
+print(net)
+
+net.save("model.pkl")
+loaded = hl.LCN.load("model.pkl")
+```
+
+### Step by step (same loop, more control)
+
+`fit` is nothing but this loop — drive it yourself to interleave your own
+metrics, schedules, or early stopping:
+
+```python
+for epoch in range(epochs):
+    net.set_epoch(epoch, epochs)          # cosine learning rate
+    for start in range(0, count, batch_size):
+        net.zero_grad()
+        for i in range(start, start + batch_size):
+            net.forward(fields[i])
+            net.loss(targets[i])          # seeds the gradient
+            net.backward()                # accumulates (sums) it
+        net.adam()                        # one step per batch
+    net.observe(epoch_metric, epoch)      # restore_best bookkeeping
+net.restore_best()
+```
+
+The gradient is a **sum** over the batch, so the effective step scales with
+batch size — the same convention as the C++ examples.
+
+---
+
+## Features
+
+- **One class** — `hypercube_lcn.LCN` is the whole product surface
+- **`fit`** — shuffle, batch, cosine schedule, restore-best, in one call
+- **Custom loops** — `zero_grad` / `forward` / `loss` / `backward` / `adam` /
+  `set_epoch` exposed one-to-one with the C++ API
+- **dim 4–24** — field length N = 2<sup>dim</sup>; depth `z_max`
+  (default: dim); lookback window `gather_span` 2–6
+- **Masked loss** — targets narrower than N leave the rest of the cube as
+  free hidden units
+- **`tanh_last`** — off by default (raw accumulator out); on confines the
+  output to (-1, 1)
+- **Weights and gradient as NumPy** — `net.weights` (settable) and
+  `net.grad`, z-major layout: depth, vertex, axis, tap
+- **Save / load** — `save` / `load` (pickle: config + weights; optimizer
+  state is not stored)
+- **NumPy float32** — arrays converted for you; prefer contiguous float32
+
+---
+
+## Examples
+
+For a first try, paste the [Quick start](#quick-start) after
+`pip install hypercube-lcn`. That is self-contained.
+
+The demo scripts on GitHub under
+[`python/examples/`](https://github.com/dliptak001/HypercubeLCN/tree/main/python/examples)
+are there to open or download — they are not added to your machine by pip.
+
+| Script | What it is for |
+|--------|----------------|
+| [synthetic_regression.py](https://github.com/dliptak001/HypercubeLCN/blob/main/python/examples/synthetic_regression.py) | Toy field map: `fit` with verbose loss, then held-out MSE |
+
+```bash
+# from a clone of HypercubeLCN, after: pip install hypercube-lcn
+python python/examples/synthetic_regression.py
+```
+
+These use easy made-up fields so the API is obvious — not scores to publish.
+
+---
+
+## Documentation
+
+| Doc | Role |
+|-----|------|
+| **[docs/Python_SDK.md](https://github.com/dliptak001/HypercubeLCN/blob/main/docs/Python_SDK.md)** | Canonical Python API — every method, layout, pickle, limits |
+| [docs/CPP_SDK.md](https://github.com/dliptak001/HypercubeLCN/blob/main/docs/CPP_SDK.md) | Native library guide (same product, C++) |
+| [Project README](https://github.com/dliptak001/HypercubeLCN#readme) | Product story and C++ demos from the repo root |
+| [docs/forward.md](https://github.com/dliptak001/HypercubeLCN/blob/main/docs/forward.md) | The forward pass, loop by loop, with hand-checkable examples |
+| [docs/training.md](https://github.com/dliptak001/HypercubeLCN/blob/main/docs/training.md) | Backprop and Adam through the same loops |
+| [examples/README.md](https://github.com/dliptak001/HypercubeLCN/blob/main/examples/README.md) | The C++ example programs and datasets |
+
+---
+
+## Ecosystem
+
+- **[HypercubeEtalon](https://github.com/dliptak001/HypercubeEtalon)** — frozen etalon transit + thin readout.
+- **[HypercubeWTF](https://github.com/dliptak001/HypercubeWTF)** — frozen reservoir orbit + thin readout.
+- **[HypercubeCascade](https://github.com/dliptak001/HypercubeCascade)** — both frozen stages in series + thin readout.
+- **[HypercubeCNN](https://github.com/dliptak001/HypercubeCNN)** — cube-native conv stack with shared kernels.
+- **[HypercubeESN](https://github.com/dliptak001/HypercubeESN)** — echo-state / reservoir computing on streams.
+- **[HypercubeHopfield](https://github.com/dliptak001/HypercubeHopfield)** — Hopfield-style dynamics on the cube.
+
+---
+
+## License
+
+Apache 2.0.