cid-engine 0.4.2__tar.gz

cid_engine-0.4.2/LICENSE

@@ -0,0 +1,202 @@
+
+                                 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 [yyyy] [name of copyright owner]
+
+   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.

cid_engine-0.4.2/MANIFEST.in

@@ -0,0 +1,4 @@
+include LICENSE
+include README.md
+recursive-include csrc *.cpp *.cu *.cuh *.h *.hpp
+recursive-include src *.py

cid_engine-0.4.2/PKG-INFO

@@ -0,0 +1,175 @@
+Metadata-Version: 2.4
+Name: cid-engine
+Version: 0.4.2
+Summary: C++20 execution engine for Continuous Interaction Diffusion
+Author: fwerkor
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/fwerkor/cid-engine
+Project-URL: Repository, https://github.com/fwerkor/cid-engine
+Project-URL: Issues, https://github.com/fwerkor/cid-engine/issues
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Science/Research
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Programming Language :: C++
+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: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: torch>=2.4
+Provides-Extra: dev
+Requires-Dist: hypothesis>=6.112; extra == "dev"
+Requires-Dist: pytest>=8; extra == "dev"
+Requires-Dist: ruff>=0.8; extra == "dev"
+Provides-Extra: release
+Requires-Dist: build>=1.2; extra == "release"
+Requires-Dist: twine>=6; extra == "release"
+Dynamic: license-file
+
+# cid-engine
+
+cid-engine is the native execution engine for Continuous Interaction Diffusion (CID).
+
+The engine core is written in C++20. Python is a thin integration surface for the current CID
+model and for semantic-reference tests; it is not the execution core.
+
+## Install
+
+PyPI releases can be installed with:
+
+    pip install cid-engine
+
+The current PyPI distribution is source-based so that installation can select the correct native
+backend for the target machine. A C++20 compiler is required. If the installed PyTorch build has
+CUDA support and `CUDA_HOME` points to a CUDA toolkit, the CUDA backend is built automatically;
+otherwise cid-engine builds its CPU backend.
+
+For an environment that already pins a particular PyTorch build, install against that exact local
+PyTorch ABI with:
+
+    pip install cid-engine --no-build-isolation
+
+## Architecture
+
+    CID Python/model frontend
+              |
+              v
+       thin Python API
+              |
+              v
+    +-----------------------+
+    | cid-engine C++20 core |
+    | ATen custom operators |
+    +-----------------------+
+              |
+              +---- CPU
+              +---- CUDA       (display, allocation, and materialization kernels)
+              +---- Ascend     (planned)
+
+The current C++ core owns CID-specific tensor primitives and the post-statistics display refinement policy:
+
+- live thought-slot occupancy with retired-slot masking;
+- deterministic first-free prefix allocation;
+- thought diffusion corruption from a supplied epsilon tensor;
+- display-token confidence/prediction statistics;
+- C++ reveal/revision/EOS/structural-edit policy from those statistics;
+- compact materialization snapshots for one-transfer TCT control-state decoding.
+
+They are registered as torch.ops.cid_engine C++ operators, so tensors cross the Python/C++
+boundary without NumPy copies. The pure-Python implementation remains only as a semantic oracle.
+
+## Semantic contract
+
+The default engine may change floating-point reduction order, but it must not change CID
+algorithmic decisions or runtime policy.
+
+Tests require exact equality for boolean masks, allocation decisions, token IDs, and argmax
+results, plus numerically close floating-point outputs for mathematically equivalent reductions.
+
+Hot-path native operators assume token IDs already satisfy the model vocabulary contract; CID validates
+that invariant when constructing the display tensor, avoiding a device-to-host synchronization per step.
+
+Quantization, skipped diffusion steps, approximate attention, speculative execution, or new
+early-exit policies are outside this contract.
+
+## Build
+
+A C++20 compiler and PyTorch/LibTorch are required.
+
+    pip install torch
+    pip install -e '.[dev]' --no-build-isolation
+    pytest
+    ruff check .
+
+The core can also be built and tested without the Python API:
+
+    cmake -S . -B build -DCMAKE_PREFIX_PATH="$(python -c 'import torch; print(torch.utils.cmake_prefix_path)')"
+    cmake --build build
+    ctest --test-dir build --output-on-failure
+
+## Verification
+
+`cid-engine` treats semantic and numerical equivalence as release requirements. The test stack has
+three complementary layers:
+
+- property-based differential fuzzing compares native operators with independent Python semantic
+  oracles across randomized shapes, dtypes, strides, thresholds, EOS layouts, and structural edits;
+- precision tests compare floating-point kernels with float64 oracles on adversarial logits,
+  near-ties, large offsets, diffusion boundary timesteps, and vocabulary sizes up to 65,537;
+- a Clang libFuzzer target drives the C++ core directly under AddressSanitizer and
+  UndefinedBehaviorSanitizer.
+
+The normal test suite uses a bounded fuzz budget. A scheduled stress workflow raises the Hypothesis
+budget substantially and runs the coverage-guided native fuzzer. The same native target can be run
+locally with Clang:
+
+    CXX=clang++ cmake -S . -B build-fuzz \
+      -DCID_ENGINE_BUILD_TESTS=OFF -DCID_ENGINE_BUILD_FUZZERS=ON \
+      -DCMAKE_PREFIX_PATH="$(python -c 'import torch; print(torch.utils.cmake_prefix_path)')"
+    cmake --build build-fuzz --parallel 2
+    cp -a tests/fuzz/corpus /tmp/cid-engine-fuzz-corpus
+    OMP_NUM_THREADS=1 MKL_NUM_THREADS=1 ASAN_OPTIONS=detect_leaks=0 \
+      ./build-fuzz/cid_engine_fuzz /tmp/cid-engine-fuzz-corpus -max_total_time=60
+
+## Use
+
+    import torch
+    import cid_engine
+
+    token_ids = torch.randint(0, 32000, (1, 128), device="cuda")
+    logits = torch.randn(1, 128, 32000, device="cuda", dtype=torch.bfloat16)
+
+    confidence, predicted, current_confidence = cid_engine.display_token_statistics(
+        token_ids,
+        logits,
+    )
+
+The underlying operator is also available directly through
+torch.ops.cid_engine.display_token_statistics. The Python reference module is retained only for semantic tests and benchmarks.
+
+## Benchmark
+
+    cid-engine-bench --device cuda --batch 1 --tokens 128 --vocab 65536
+
+The benchmark verifies C++/reference equivalence before reporting timing.
+
+Moving Python tensor expressions into C++ does not by itself guarantee a speedup. If C++ launches
+the same sequence of ATen kernels, device work is essentially unchanged. Establishing the native
+core first gives CID a stable place for fused CUDA, CPU, and Ascend implementations and for future
+scheduler and memory-planner logic.
+
+## Roadmap
+
+1. Keep model-visible TCT numeric state device-resident across CID steps.
+2. Move the remaining materialization/control-state transitions into native state.
+3. Expand fused CUDA coverage only where profiling shows a material gain.
+4. Add CID-aware buffer lifetime and stream scheduling.
+5. Overlap model/device work with asynchronous tool/source execution.
+6. Add the Ascend backend behind the same C++ engine interface.
+7. Keep PyTorch as the training/reference frontend until replacing a layer has measured value.
+
+Apache-2.0, matching the main CID repository.

cid_engine-0.4.2/README.md

@@ -0,0 +1,143 @@
+# cid-engine
+
+cid-engine is the native execution engine for Continuous Interaction Diffusion (CID).
+
+The engine core is written in C++20. Python is a thin integration surface for the current CID
+model and for semantic-reference tests; it is not the execution core.
+
+## Install
+
+PyPI releases can be installed with:
+
+    pip install cid-engine
+
+The current PyPI distribution is source-based so that installation can select the correct native
+backend for the target machine. A C++20 compiler is required. If the installed PyTorch build has
+CUDA support and `CUDA_HOME` points to a CUDA toolkit, the CUDA backend is built automatically;
+otherwise cid-engine builds its CPU backend.
+
+For an environment that already pins a particular PyTorch build, install against that exact local
+PyTorch ABI with:
+
+    pip install cid-engine --no-build-isolation
+
+## Architecture
+
+    CID Python/model frontend
+              |
+              v
+       thin Python API
+              |
+              v
+    +-----------------------+
+    | cid-engine C++20 core |
+    | ATen custom operators |
+    +-----------------------+
+              |
+              +---- CPU
+              +---- CUDA       (display, allocation, and materialization kernels)
+              +---- Ascend     (planned)
+
+The current C++ core owns CID-specific tensor primitives and the post-statistics display refinement policy:
+
+- live thought-slot occupancy with retired-slot masking;
+- deterministic first-free prefix allocation;
+- thought diffusion corruption from a supplied epsilon tensor;
+- display-token confidence/prediction statistics;
+- C++ reveal/revision/EOS/structural-edit policy from those statistics;
+- compact materialization snapshots for one-transfer TCT control-state decoding.
+
+They are registered as torch.ops.cid_engine C++ operators, so tensors cross the Python/C++
+boundary without NumPy copies. The pure-Python implementation remains only as a semantic oracle.
+
+## Semantic contract
+
+The default engine may change floating-point reduction order, but it must not change CID
+algorithmic decisions or runtime policy.
+
+Tests require exact equality for boolean masks, allocation decisions, token IDs, and argmax
+results, plus numerically close floating-point outputs for mathematically equivalent reductions.
+
+Hot-path native operators assume token IDs already satisfy the model vocabulary contract; CID validates
+that invariant when constructing the display tensor, avoiding a device-to-host synchronization per step.
+
+Quantization, skipped diffusion steps, approximate attention, speculative execution, or new
+early-exit policies are outside this contract.
+
+## Build
+
+A C++20 compiler and PyTorch/LibTorch are required.
+
+    pip install torch
+    pip install -e '.[dev]' --no-build-isolation
+    pytest
+    ruff check .
+
+The core can also be built and tested without the Python API:
+
+    cmake -S . -B build -DCMAKE_PREFIX_PATH="$(python -c 'import torch; print(torch.utils.cmake_prefix_path)')"
+    cmake --build build
+    ctest --test-dir build --output-on-failure
+
+## Verification
+
+`cid-engine` treats semantic and numerical equivalence as release requirements. The test stack has
+three complementary layers:
+
+- property-based differential fuzzing compares native operators with independent Python semantic
+  oracles across randomized shapes, dtypes, strides, thresholds, EOS layouts, and structural edits;
+- precision tests compare floating-point kernels with float64 oracles on adversarial logits,
+  near-ties, large offsets, diffusion boundary timesteps, and vocabulary sizes up to 65,537;
+- a Clang libFuzzer target drives the C++ core directly under AddressSanitizer and
+  UndefinedBehaviorSanitizer.
+
+The normal test suite uses a bounded fuzz budget. A scheduled stress workflow raises the Hypothesis
+budget substantially and runs the coverage-guided native fuzzer. The same native target can be run
+locally with Clang:
+
+    CXX=clang++ cmake -S . -B build-fuzz \
+      -DCID_ENGINE_BUILD_TESTS=OFF -DCID_ENGINE_BUILD_FUZZERS=ON \
+      -DCMAKE_PREFIX_PATH="$(python -c 'import torch; print(torch.utils.cmake_prefix_path)')"
+    cmake --build build-fuzz --parallel 2
+    cp -a tests/fuzz/corpus /tmp/cid-engine-fuzz-corpus
+    OMP_NUM_THREADS=1 MKL_NUM_THREADS=1 ASAN_OPTIONS=detect_leaks=0 \
+      ./build-fuzz/cid_engine_fuzz /tmp/cid-engine-fuzz-corpus -max_total_time=60
+
+## Use
+
+    import torch
+    import cid_engine
+
+    token_ids = torch.randint(0, 32000, (1, 128), device="cuda")
+    logits = torch.randn(1, 128, 32000, device="cuda", dtype=torch.bfloat16)
+
+    confidence, predicted, current_confidence = cid_engine.display_token_statistics(
+        token_ids,
+        logits,
+    )
+
+The underlying operator is also available directly through
+torch.ops.cid_engine.display_token_statistics. The Python reference module is retained only for semantic tests and benchmarks.
+
+## Benchmark
+
+    cid-engine-bench --device cuda --batch 1 --tokens 128 --vocab 65536
+
+The benchmark verifies C++/reference equivalence before reporting timing.
+
+Moving Python tensor expressions into C++ does not by itself guarantee a speedup. If C++ launches
+the same sequence of ATen kernels, device work is essentially unchanged. Establishing the native
+core first gives CID a stable place for fused CUDA, CPU, and Ascend implementations and for future
+scheduler and memory-planner logic.
+
+## Roadmap
+
+1. Keep model-visible TCT numeric state device-resident across CID steps.
+2. Move the remaining materialization/control-state transitions into native state.
+3. Expand fused CUDA coverage only where profiling shows a material gain.
+4. Add CID-aware buffer lifetime and stream scheduling.
+5. Overlap model/device work with asynchronous tool/source execution.
+6. Add the Ascend backend behind the same C++ engine interface.
+7. Keep PyTorch as the training/reference frontend until replacing a layer has measured value.
+
+Apache-2.0, matching the main CID repository.

cid_engine-0.4.2/csrc/bindings.cpp

@@ -0,0 +1,10 @@
+#include <pybind11/pybind11.h>
+
+PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
+  m.doc() = "Native C++20 core for cid-engine";
+#ifdef CID_ENGINE_WITH_CUDA
+  m.attr("cuda_backend_built") = pybind11::bool_(true);
+#else
+  m.attr("cuda_backend_built") = pybind11::bool_(false);
+#endif
+}