sequenzo 0.1.18__tar.gz → 0.1.20__tar.gz

{sequenzo-0.1.18 → sequenzo-0.1.20}/MANIFEST.in

@@ -11,10 +11,8 @@ recursive-include sequenzo/clustering/src *.cpp
 recursive-include sequenzo/clustering/src *.h
 recursive-include sequenzo/clustering/src *.hpp
 
-# Include all C source files
-recursive-include sequenzo/dissimilarity_measures/utils *.c
-recursive-include sequenzo/clustering/utils *.c
-recursive-include sequenzo/big_data/clara/utils *.c
+# Note: .c files are auto-generated from .pyx during build, so we don't include them here
+# Only .pyx source files are needed for distribution
 
 # Include all header files and external libraries
 recursive-include sequenzo/dissimilarity_measures/src/xsimd/include *.hpp

{sequenzo-0.1.18/sequenzo.egg-info → sequenzo-0.1.20}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sequenzo
-Version: 0.1.18
+Version: 0.1.20
 Summary: A fast, scalable and intuitive Python package for social sequence analysis.
 Author-email: Yuqi Liang <yuqi.liang.1900@gmail.com>, Xinyi Li <1836724126@qq.com>, Jan Heinrich Ernst Meyerhoff-Liang <jan.meyerhoff1@gmail.com>
 License: BSD 3-Clause License
@@ -54,11 +54,15 @@ Requires-Dist: cython>=0.29.21
 Requires-Dist: scipy>=1.6.3
 Requires-Dist: scikit-learn>=0.24.2
 Requires-Dist: fastcluster>=1.2.6
+Requires-Dist: rpy2>=3.5.12; python_version >= "3.12"
+Requires-Dist: rpy2>=3.5.6; python_version == "3.11"
+Requires-Dist: rpy2>=3.5.6; python_version == "3.10"
+Requires-Dist: rpy2>=3.5.6; python_version == "3.9"
 Requires-Dist: joblib>=1.0.1
 Requires-Dist: docutils>=0.17
 Requires-Dist: tqdm<5.0.0,>=4.62.3
 Requires-Dist: missingno<0.6.0,>=0.5.2
-Requires-Dist: rpy2
+Requires-Dist: cffi>=1.15.0
 Provides-Extra: dev
 Requires-Dist: pytest>=6.2.5; extra == "dev"
 Requires-Dist: flake8>=3.9.2; extra == "dev"
@@ -97,6 +101,11 @@ Sequenzo outperforms traditional R-based tools in social sequence analysis, deli
 >  
 > 📖 Available in **English and Chinese**, our docs are written to be approachable, practical, and easy to follow.
 
+## ✨ Be part of the Sequenzo community
+Join our Discord channel to iscuss ideas, get help, and hear about upcoming Sequenzo versions, tutorials, and workshops first.
+
+➡️ https://discord.gg/3bMDKRHW
+
 ## Target Users
 
 Sequenzo is designed for:
@@ -172,6 +181,50 @@ If you have some issues with the installation, it might because you have both Py
 pip3 install sequenzo
 ```
 
+### ⚠️ Having Installation or Import Issues?
+
+**Error:** `ImportError: numpy.core.multiarray failed to import`
+
+**Most likely cause:** NumPy version mismatch (you have NumPy 1.x, but need 2.x)
+
+**Quick Fix** (copy-paste these commands):
+```bash
+# Check your NumPy version first
+python -c "import numpy; print(f'NumPy: {numpy.__version__}')"
+
+# If you see 1.x.x, upgrade to 2.x:
+pip install --upgrade "numpy>=2.0.0"
+pip uninstall sequenzo -y
+pip install --no-cache-dir sequenzo
+```
+
+📖 **Still having issues?** 
+1. Run our diagnostic tool to identify the problem:
+   ```bash
+   curl -O https://raw.githubusercontent.com/Liang-Team/Sequenzo/main/diagnose.py
+   python diagnose.py
+   ```
+2. See our detailed guides:
+   - **[QUICK_FIX.md](QUICK_FIX.md)** - Simple step-by-step solutions
+   - **[TROUBLESHOOTING.md](TROUBLESHOOTING.md)** - Comprehensive troubleshooting
+   - **[docs/WHY_IMPORT_FAILS.md](docs/WHY_IMPORT_FAILS.md)** - Technical explanation
+
+### Optional R Integration
+
+Sequenzo now checks the system environment variables before running ward.D hierarchical clustering.
+
+If R is missing, a relevant prompt will be displayed along with specific installation instructions. If `fastcluster` is missing, Sequenzo will automatically download `fastcluster`.
+
+Before automatically downloading `fastcluster`, Sequenzo checks whether R is available; if R is not installed, sequenzo will not automatically download fastcluster.
+
+Sequenzo supports advanced Ward clustering methods that require R integration. If you need to use the `ward_d` clustering method, install with R support:
+
+```
+pip install sequenzo[r]
+```
+
+This will install the optional `rpy2` dependency, which provides Python-R interoperability. Note that R must also be installed on your system for `rpy2` to work.
+
 For more information about the latest stable release and required dependencies, please refer to [PyPI](https://pypi.org/project/sequenzo/). 
 
 ## Documentation
@@ -209,7 +262,7 @@ Check out our [contribution guide]() for more details (work in progress).
 If you're contributing or debugging, use:
 
 ```bash
-pip install -r requirements-3.10.txt  # Or matching your Python version
+pip install -r requirements/requirements-3.10.txt  # Or matching your Python version
 ```
     
 For standard installation, use:
@@ -222,23 +275,31 @@ pip install .  # Uses pyproject.toml
 
 **Paper Authors**
 * [Yuqi Liang, University of Oxford](https://www.yuqi-liang.tech/)
-* [Xinyi Li, Heilongjiang University](https://github.com/Fantasy201)
+* [Xinyi Li, Northeastern University](https://github.com/Fantasy201)
 * [Jan Heinrich Ernst Meyerhoff-Liang, Institute for New Economic Thinking Oxford](https://www.inet.ox.ac.uk/people/jan-meyerhoff-liang)
 
 **Package Contributors**
-* [Liangxingyun He, Stockholm School of Economics (Sweden)](https://www.linkedin.com/in/liangxingyun-he-6aa128304/)
-* [Yukun Ming, Universidad Carlos III de Madrid (Spain)](www.linkedin.com/in/yukun)
-* [Ziting Yang, Rochester Wniversity (US)](www.linkedin.com/in/ziting-yang-7b33832bb)
+
+Coding contributors:
+* [Sebastian Daza](https://sdaza.com/)
+* [Cheng Deng](https://github.com/de-de-de-de-de)
+* [Liangxingyun He, Stockholm School of Economics, Sweden](https://www.linkedin.com/in/liangxingyun-he-6aa128304/)
+
+Documentation contributors:
+* [Liangxingyun He, Stockholm School of Economics, Sweden](https://www.linkedin.com/in/liangxingyun-he-6aa128304/)
+* [Yukun Ming, Universidad Carlos III de Madrid (Spain)](https://www.linkedin.com/in/yukun)
 * [Sizhu Qu, Northeastern University (US)](https://www.linkedin.com/in/sizhuq)
+* [Ziting Yang, Rochester Wniversity (US)](https://www.linkedin.com/in/ziting-yang-7b33832bb)
+
+Others
 * With special thanks to our initial testers (alphabetically ordered): [Joji Chia](https://sociology.illinois.edu/directory/profile/jbchia2), [Kass Gonzalez](https://www.linkedin.com/in/kass-gonzalez-72a778276/), [Sinyee Lu](https://sociology.illinois.edu/directory/profile/qianyil4), [Sohee Shin](https://sociology.illinois.edu/directory/profile/sohees2)
 * Website and related technical support: [Mactavish](https://github.com/mactavishz)
-* PAM clustering algorithm implementation: Cheng Deng
 * Sequence data sources compilation - History: Jingrui Chen
 * Visual design consultant: Changyu Yi
 
 **Acknowledgements**
 
-* Methodological advisor in sequence analysis: [Professor Tim Liao, University of Illinois Urbana-Champaign](https://sociology.illinois.edu/directory/profile/tfliao)
-* Yuqi's PhD advisor: [Professor Ridhi Kashyap, University of Oxford](https://www.nuffield.ox.ac.uk/people/profiles/ridhi-kashyap/)
+* Methodological advisor in sequence analysis: [Professor Tim Liao (University of Illinois Urbana-Champaign)](https://sociology.illinois.edu/directory/profile/tfliao)
+* Yuqi's PhD advisor [Professor Ridhi Kashyap (University of Oxford)](https://www.nuffield.ox.ac.uk/people/profiles/ridhi-kashyap/), and mentor [Charles Rahal (University of Oxford)](https://crahal.com/)
 * Yuqi's original programming mentor: [JiangHuShiNian](https://github.com/jianghushinian)
 

{sequenzo-0.1.18 → sequenzo-0.1.20}/README.md

@@ -31,6 +31,11 @@ Sequenzo outperforms traditional R-based tools in social sequence analysis, deli
 >  
 > 📖 Available in **English and Chinese**, our docs are written to be approachable, practical, and easy to follow.
 
+## ✨ Be part of the Sequenzo community
+Join our Discord channel to iscuss ideas, get help, and hear about upcoming Sequenzo versions, tutorials, and workshops first.
+
+➡️ https://discord.gg/3bMDKRHW
+
 ## Target Users
 
 Sequenzo is designed for:
@@ -106,6 +111,50 @@ If you have some issues with the installation, it might because you have both Py
 pip3 install sequenzo
 ```
 
+### ⚠️ Having Installation or Import Issues?
+
+**Error:** `ImportError: numpy.core.multiarray failed to import`
+
+**Most likely cause:** NumPy version mismatch (you have NumPy 1.x, but need 2.x)
+
+**Quick Fix** (copy-paste these commands):
+```bash
+# Check your NumPy version first
+python -c "import numpy; print(f'NumPy: {numpy.__version__}')"
+
+# If you see 1.x.x, upgrade to 2.x:
+pip install --upgrade "numpy>=2.0.0"
+pip uninstall sequenzo -y
+pip install --no-cache-dir sequenzo
+```
+
+📖 **Still having issues?** 
+1. Run our diagnostic tool to identify the problem:
+   ```bash
+   curl -O https://raw.githubusercontent.com/Liang-Team/Sequenzo/main/diagnose.py
+   python diagnose.py
+   ```
+2. See our detailed guides:
+   - **[QUICK_FIX.md](QUICK_FIX.md)** - Simple step-by-step solutions
+   - **[TROUBLESHOOTING.md](TROUBLESHOOTING.md)** - Comprehensive troubleshooting
+   - **[docs/WHY_IMPORT_FAILS.md](docs/WHY_IMPORT_FAILS.md)** - Technical explanation
+
+### Optional R Integration
+
+Sequenzo now checks the system environment variables before running ward.D hierarchical clustering.
+
+If R is missing, a relevant prompt will be displayed along with specific installation instructions. If `fastcluster` is missing, Sequenzo will automatically download `fastcluster`.
+
+Before automatically downloading `fastcluster`, Sequenzo checks whether R is available; if R is not installed, sequenzo will not automatically download fastcluster.
+
+Sequenzo supports advanced Ward clustering methods that require R integration. If you need to use the `ward_d` clustering method, install with R support:
+
+```
+pip install sequenzo[r]
+```
+
+This will install the optional `rpy2` dependency, which provides Python-R interoperability. Note that R must also be installed on your system for `rpy2` to work.
+
 For more information about the latest stable release and required dependencies, please refer to [PyPI](https://pypi.org/project/sequenzo/). 
 
 ## Documentation
@@ -143,7 +192,7 @@ Check out our [contribution guide]() for more details (work in progress).
 If you're contributing or debugging, use:
 
 ```bash
-pip install -r requirements-3.10.txt  # Or matching your Python version
+pip install -r requirements/requirements-3.10.txt  # Or matching your Python version
 ```
     
 For standard installation, use:
@@ -156,23 +205,31 @@ pip install .  # Uses pyproject.toml
 
 **Paper Authors**
 * [Yuqi Liang, University of Oxford](https://www.yuqi-liang.tech/)
-* [Xinyi Li, Heilongjiang University](https://github.com/Fantasy201)
+* [Xinyi Li, Northeastern University](https://github.com/Fantasy201)
 * [Jan Heinrich Ernst Meyerhoff-Liang, Institute for New Economic Thinking Oxford](https://www.inet.ox.ac.uk/people/jan-meyerhoff-liang)
 
 **Package Contributors**
-* [Liangxingyun He, Stockholm School of Economics (Sweden)](https://www.linkedin.com/in/liangxingyun-he-6aa128304/)
-* [Yukun Ming, Universidad Carlos III de Madrid (Spain)](www.linkedin.com/in/yukun)
-* [Ziting Yang, Rochester Wniversity (US)](www.linkedin.com/in/ziting-yang-7b33832bb)
+
+Coding contributors:
+* [Sebastian Daza](https://sdaza.com/)
+* [Cheng Deng](https://github.com/de-de-de-de-de)
+* [Liangxingyun He, Stockholm School of Economics, Sweden](https://www.linkedin.com/in/liangxingyun-he-6aa128304/)
+
+Documentation contributors:
+* [Liangxingyun He, Stockholm School of Economics, Sweden](https://www.linkedin.com/in/liangxingyun-he-6aa128304/)
+* [Yukun Ming, Universidad Carlos III de Madrid (Spain)](https://www.linkedin.com/in/yukun)
 * [Sizhu Qu, Northeastern University (US)](https://www.linkedin.com/in/sizhuq)
+* [Ziting Yang, Rochester Wniversity (US)](https://www.linkedin.com/in/ziting-yang-7b33832bb)
+
+Others
 * With special thanks to our initial testers (alphabetically ordered): [Joji Chia](https://sociology.illinois.edu/directory/profile/jbchia2), [Kass Gonzalez](https://www.linkedin.com/in/kass-gonzalez-72a778276/), [Sinyee Lu](https://sociology.illinois.edu/directory/profile/qianyil4), [Sohee Shin](https://sociology.illinois.edu/directory/profile/sohees2)
 * Website and related technical support: [Mactavish](https://github.com/mactavishz)
-* PAM clustering algorithm implementation: Cheng Deng
 * Sequence data sources compilation - History: Jingrui Chen
 * Visual design consultant: Changyu Yi
 
 **Acknowledgements**
 
-* Methodological advisor in sequence analysis: [Professor Tim Liao, University of Illinois Urbana-Champaign](https://sociology.illinois.edu/directory/profile/tfliao)
-* Yuqi's PhD advisor: [Professor Ridhi Kashyap, University of Oxford](https://www.nuffield.ox.ac.uk/people/profiles/ridhi-kashyap/)
+* Methodological advisor in sequence analysis: [Professor Tim Liao (University of Illinois Urbana-Champaign)](https://sociology.illinois.edu/directory/profile/tfliao)
+* Yuqi's PhD advisor [Professor Ridhi Kashyap (University of Oxford)](https://www.nuffield.ox.ac.uk/people/profiles/ridhi-kashyap/), and mentor [Charles Rahal (University of Oxford)](https://crahal.com/)
 * Yuqi's original programming mentor: [JiangHuShiNian](https://github.com/jianghushinian)
 

{sequenzo-0.1.18 → sequenzo-0.1.20}/pyproject.toml

@@ -29,7 +29,7 @@ classifiers = [
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
 ]
-version = "0.1.18"
+version = "0.1.20"
 dependencies = [
     "numpy>=1.19.5,<2.5",
     "pandas>=1.2.5",
@@ -41,11 +41,17 @@ dependencies = [
     "scipy>=1.6.3",
     "scikit-learn>=0.24.2",
     "fastcluster>=1.2.6",
+    # rpy2 is required at runtime for Ward D via R fastcluster.
+    # Use PEP 508 markers to select versions per Python runtime.
+    "rpy2>=3.5.12; python_version >= '3.12'",
+    "rpy2>=3.5.6; python_version == '3.11'",
+    "rpy2>=3.5.6; python_version == '3.10'",
+    "rpy2>=3.5.6; python_version == '3.9'",
     "joblib>=1.0.1",
     "docutils>=0.17",
     "tqdm>=4.62.3,<5.0.0",
     "missingno>=0.5.2,<0.6.0",
-    "rpy2",
+    "cffi>=1.15.0",  # Required for C extensions and FFI functionality
 ]
 
 [project.optional-dependencies]
@@ -66,17 +72,13 @@ include-package-data = true
 "sequenzo.dissimilarity_measures" = [
   "src/**/*.cpp",
   "src/**/*.h", 
-  "src/**/*.hpp",
-  "utils/*.c"
+  "src/**/*.hpp"
 ]
 "sequenzo.clustering" = [
   "src/**/*.cpp",
-  "src/**/*.h",
-  "utils/*.c"
-]
-"sequenzo.big_data.clara" = [
-  "utils/*.c"
+  "src/**/*.h"
 ]
+# Note: .c files are auto-generated from .pyx during build via Cython
 
 
 [tool.setuptools.packages.find]

{sequenzo-0.1.18 → sequenzo-0.1.20}/sequenzo/__init__.py

@@ -4,13 +4,6 @@
 @Time    : 11/02/2025 16:41
 @Desc    : 
 """
-# sequenzo/sequenzo/__init__.py (Inside the inner sequenzo package)
-# This file is now needed for making the submodules accessible.
-
-# You *don't* need to handle the Rust extension here.
-# You can have package level docstring here.
-# No need for __all__ here, use relative import in the top-level __init__.py
-
 from .datasets import load_dataset, list_datasets
 
 # Import the core functions that should be directly available from the sequenzo package
@@ -206,3 +199,42 @@ __all__ = [
     # Keep old names for backward compatibility
     "seqsamm"
 ]
+
+# Auto-setup OpenMP on Apple Silicon (runs on import)
+def _setup_openmp_if_needed():
+    """
+    Automatically setup OpenMP dependencies on Apple Silicon Macs.
+    This function runs on import to ensure OpenMP is available.
+    """
+    import sys
+    import os
+    import platform
+    
+    # Only run on macOS
+    if sys.platform != 'darwin':
+        return
+    
+    # Only run on Apple Silicon
+    if platform.machine() != 'arm64':
+        return
+    
+    # Check if we're in a conda environment (don't interfere)
+    if os.environ.get('CONDA_DEFAULT_ENV'):
+        return
+    
+    # Try to import and run the OpenMP setup
+    try:
+        from .openmp_setup import ensure_openmp_support
+        ensure_openmp_support()
+    except ImportError:
+        # OpenMP setup module not available, skip
+        pass
+    except Exception:
+        # Any other error, skip silently
+        pass
+
+# Run the setup function
+_setup_openmp_if_needed()
+
+# Clean up the setup function from the module namespace
+del _setup_openmp_if_needed