syntharc 0.1.0__tar.gz

syntharc-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Fahad Abdullah
+
+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.

syntharc-0.1.0/PKG-INFO

@@ -0,0 +1,243 @@
+Metadata-Version: 2.4
+Name: syntharc
+Version: 0.1.0
+Summary: Unified synthetic data generation for tabular, time-series, image, and text data
+Author-email: Fahad Abdullah <fahadai.co@gmail.com>
+License: MIT License
+        
+        Copyright (c) 2026 Fahad Abdullah
+        
+        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.
+        
+Project-URL: Homepage, https://github.com/FAbdullah17/syntharc
+Project-URL: Repository, https://github.com/FAbdullah17/syntharc
+Project-URL: Issues, https://github.com/FAbdullah17/syntharc/issues
+Project-URL: Documentation, https://github.com/FAbdullah17/syntharc/tree/main/docs
+Keywords: synthetic-data,syntharceration,augmentation,tabular,time-series,image,text
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pandas>=1.5
+Requires-Dist: numpy>=1.23
+Requires-Dist: scikit-learn>=1.2
+Requires-Dist: matplotlib>=3.6
+Requires-Dist: Pillow>=9.0
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: rich>=13.0
+Requires-Dist: torch>=2.0
+Provides-Extra: tabular
+Requires-Dist: sdv>=1.10; extra == "tabular"
+Provides-Extra: timeseries
+Requires-Dist: sdv>=1.10; extra == "timeseries"
+Provides-Extra: image
+Requires-Dist: albumentations>=1.4; extra == "image"
+Requires-Dist: opencv-python-headless>=4.9; extra == "image"
+Provides-Extra: text
+Requires-Dist: transformers>=4.40; extra == "text"
+Requires-Dist: accelerate>=0.22; extra == "text"
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == "dev"
+Requires-Dist: pytest-cov>=5.0; extra == "dev"
+Requires-Dist: ruff>=0.4; extra == "dev"
+Requires-Dist: mypy>=1.10; extra == "dev"
+Requires-Dist: pre-commit>=3.7; extra == "dev"
+Requires-Dist: pandas-stubs; extra == "dev"
+Requires-Dist: types-PyYAML; extra == "dev"
+Requires-Dist: types-Pillow; extra == "dev"
+Provides-Extra: all
+Requires-Dist: syntharc[tabular]; extra == "all"
+Requires-Dist: syntharc[timeseries]; extra == "all"
+Requires-Dist: syntharc[image]; extra == "all"
+Requires-Dist: syntharc[text]; extra == "all"
+Requires-Dist: syntharc[dev]; extra == "all"
+Dynamic: license-file
+
+# syntharc: The Grand Unified Synthetic Data Generator
+
+[![CI](https://github.com/your-username/syntharc/actions/workflows/ci.yml/badge.svg)](https://github.com/your-username/syntharc/actions/workflows/ci.yml)
+[![PyPI version](https://badge.fury.io/py/syntharc.svg)](https://badge.fury.io/py/syntharc)
+[![Python Versions](https://img.shields.io/pypi/pyversions/syntharc.svg)](https://pypi.org/project/syntharc/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+
+**`syntharc`** is a powerful, flexible, and unified Python library for generating high-quality synthetic data across multiple domains. In the modern machine learning ecosystem, acquiring high-quality, privacy-compliant, and diverse datasets is often the biggest bottleneck. `syntharc` solves this by abstracting the complexity of various generative algorithms into a single, intuitive framework.
+
+Whether you need to anonymize sensitive tabular records, augment image operations for computer vision, simulate realistic text sequences, or replicate complex financial time-series signals. `syntharc` provides the unified API to do it all using state-of-the-art backend engines like PyTorch, Hugging Face Transformers, Albumentations, and the Synthetic Data Vault (SDV).
+
+---
+
+## 🧠 How It Works: The Unified Lifecycle
+
+One of the biggest hurdles in synthetic data generation is the fragmented tooling across different data types. `syntharc` forces all underlying models into a highly predictable **3-step lifecycle** inherited from our `BaseSynthesizer` architecture:
+
+1. **`prepare(**kwargs)`**: Configures the internal environment. This is where you define structural metadata, column types, sequence keys, or image transformation pipelines to guide the generation.
+2. **`fit(data)`**: Feeds your actual, real-world dataset into the generative model so it can learn patterns, probabilistic distributions, and internal representations.
+3. **`generate(**kwargs)`**: Samples from the trained model to yield your brand new synthetic dataset—retaining the statistical properties of the original without exposing real user information.
+
+---
+
+## 🚀 Key Features
+
+*   📊 **Tabular Data:** Synthesize fully relational datasets using Deep Learning (`CTGANSynthesizer`) or statistical modeling (`GaussianCopulaSynthesizer`).
+*   📝 **Text Generation:** Leverage the power of LLMs (`TransformerSynthesizer`), classic statistical chains (`MarkovSynthesizer`), or strict ruleings (`TemplateSynthesizer`).
+*   🖼️ **Image Augmentation:** Use `ImageAugmentor` to rapidly iterate through spatial and pixel-level augmentations, dynamically expanding computer vision datasets.
+*   📈 **Time-Series:** Employ the `PARSynthesizer` (Probabilistic AutoRegressive model) to safely synthesize robust sequence data over time.
+*   ✅ **Evaluations:** Built-in tools for evaluating the quality, fidelity, and privacy metrics of your generated data against your source data.
+
+---
+
+## 📦 Installation
+
+Install `syntharc` via pip. The base package provides the core infrastructure. We highly recommend installing the specific domain dependencies you intend to use to keep your environment lean:
+
+```bash
+# Install everything (Recommended for the full experience)
+pip install "syntharc[all]"
+
+# Or individually pick your domains:
+pip install "syntharc[tabular]"
+pip install "syntharc[text]"
+pip install "syntharc[image]"
+pip install "syntharc[timeseries]"
+
+# For contributors and local development:
+pip install "syntharc[dev]"
+```
+
+---
+
+## 💡 Comprehensive Quick Start
+
+Below are examples of how our unified framework elegantly handles vastly different data constraints.
+
+### 1. Privacy-Preserving Tabular Data
+Train a CTGAN model to learn the distribution of your customer data without saving real identities.
+```python
+import pandas as pd
+from syntharc.tabular.ctgan import CTGANSynthesizer
+
+real_data = pd.read_csv("customer_data.csv")
+
+# 1. Initialize & Prepare metadata
+synth = CTGANSynthesizer(epochs=50)
+synth.prepare(metadata_dict={"primary_key": "user_id"})
+
+# 2. Fit to real data
+synth.fit(real_data)
+
+# 3. Generate 1,000 synthetic rows!
+synthetic_data = synth.generate(num_rows=1000)
+print(synthetic_data.head())
+```
+
+### 2. Deep Time-Series Generation
+Synthesize sequential metrics, like stock prices or IoT sensor readings, using AutoRegressive modeling.
+```python
+from syntharc.timeseries.par import PARSynthesizer
+
+# 1. Initialize & Prepare
+synth = PARSynthesizer(epochs=25)
+# Define what makes a "sequence" (e.g., separate tracking per 'device_id')
+synth.prepare(metadata_dict={
+    "sequence_key": "device_id",
+    "context_columns": ["region"]
+})
+
+# 2. Fit
+synth.fit(sensor_dataframe)
+
+# 3. Generate sequential data
+synthetic_series = synth.generate(num_sequences=50)
+```
+
+### 3. Causal Text Generation (LLMs)
+Easily utilize Hugging Face causal language models.
+```python
+from syntharc.text.transformer import TransformerTextGenerator
+
+# 1. Initialize (Downloads SmolLM2-360M-Instruct by default)
+synth = TransformerTextGenerator()
+
+# 2. Prepare with an optional style context
+synth.prepare(corpus="The future of synthetic data is bright and highly scalable.")
+
+# 3. Generate structured text from instructions
+text_output = synth.generate(num_samples=1, instructions="Write a short summary about data.")
+print(text_output[0])
+```
+
+### 4. High-Speed Image Augmentation
+Prepare a pipeline of augmentations to expand your Machine Learning dataset effortlessly.
+```python
+from syntharc.image.augmentor import ImageAugmentor
+import cv2
+
+# 1. Initialize rules
+augmentor = ImageAugmentor()
+# Uses Albumentations backend dictionary standards mapping
+augmentor.prepare(config={
+    'resize': (256, 256),
+    'horizontal_flip': 0.5,
+    'brightness_contrast': 0.2
+})
+
+# 2. Load Real Image
+image = cv2.imread('dataset/cat.jpg')
+
+# 3. Generate augmentations
+# The generator seamlessly wraps Albumentations under the hood
+augmented_image = augmentor.generate(num_samples=1, data=[image])
+print(f"Augmented Shape: {augmented_image[0].shape}")
+```
+
+---
+
+## 📖 Documentation
+
+For comprehensive guides, parameter references, evaluations, and interactive Jupyter showcase notebooks, please check out the `docs/` folder:
+- [Getting Started](docs/getting_started.md)
+- [Tabular Generation Guide](docs/tabular.md)
+- [Time-Series Generation Guide](docs/timeseries.md)
+- [Text Generation Guide](docs/text.md)
+- [Image Augmentation Guide](docs/image.md)
+- [API Reference](docs/api_reference.md)
+
+---
+
+## 🤝 Contributing
+
+We love our contributors! If you're interested in adding a new generative model, fixing a bug, or improving the documentation, please refer to our [CONTRIBUTING.md](CONTRIBUTING.md) for detailed instructions on setting up your local environment, managing dependencies, and passing our CI checks.
+
+## 📄 License
+
+This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
+
+---
+
+Made with ❤️ by the syntharc team

syntharc-0.1.0/README.md

@@ -0,0 +1,162 @@
+# syntharc: The Grand Unified Synthetic Data Generator
+
+[![CI](https://github.com/your-username/syntharc/actions/workflows/ci.yml/badge.svg)](https://github.com/your-username/syntharc/actions/workflows/ci.yml)
+[![PyPI version](https://badge.fury.io/py/syntharc.svg)](https://badge.fury.io/py/syntharc)
+[![Python Versions](https://img.shields.io/pypi/pyversions/syntharc.svg)](https://pypi.org/project/syntharc/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+
+**`syntharc`** is a powerful, flexible, and unified Python library for generating high-quality synthetic data across multiple domains. In the modern machine learning ecosystem, acquiring high-quality, privacy-compliant, and diverse datasets is often the biggest bottleneck. `syntharc` solves this by abstracting the complexity of various generative algorithms into a single, intuitive framework.
+
+Whether you need to anonymize sensitive tabular records, augment image operations for computer vision, simulate realistic text sequences, or replicate complex financial time-series signals. `syntharc` provides the unified API to do it all using state-of-the-art backend engines like PyTorch, Hugging Face Transformers, Albumentations, and the Synthetic Data Vault (SDV).
+
+---
+
+## 🧠 How It Works: The Unified Lifecycle
+
+One of the biggest hurdles in synthetic data generation is the fragmented tooling across different data types. `syntharc` forces all underlying models into a highly predictable **3-step lifecycle** inherited from our `BaseSynthesizer` architecture:
+
+1. **`prepare(**kwargs)`**: Configures the internal environment. This is where you define structural metadata, column types, sequence keys, or image transformation pipelines to guide the generation.
+2. **`fit(data)`**: Feeds your actual, real-world dataset into the generative model so it can learn patterns, probabilistic distributions, and internal representations.
+3. **`generate(**kwargs)`**: Samples from the trained model to yield your brand new synthetic dataset—retaining the statistical properties of the original without exposing real user information.
+
+---
+
+## 🚀 Key Features
+
+*   📊 **Tabular Data:** Synthesize fully relational datasets using Deep Learning (`CTGANSynthesizer`) or statistical modeling (`GaussianCopulaSynthesizer`).
+*   📝 **Text Generation:** Leverage the power of LLMs (`TransformerSynthesizer`), classic statistical chains (`MarkovSynthesizer`), or strict ruleings (`TemplateSynthesizer`).
+*   🖼️ **Image Augmentation:** Use `ImageAugmentor` to rapidly iterate through spatial and pixel-level augmentations, dynamically expanding computer vision datasets.
+*   📈 **Time-Series:** Employ the `PARSynthesizer` (Probabilistic AutoRegressive model) to safely synthesize robust sequence data over time.
+*   ✅ **Evaluations:** Built-in tools for evaluating the quality, fidelity, and privacy metrics of your generated data against your source data.
+
+---
+
+## 📦 Installation
+
+Install `syntharc` via pip. The base package provides the core infrastructure. We highly recommend installing the specific domain dependencies you intend to use to keep your environment lean:
+
+```bash
+# Install everything (Recommended for the full experience)
+pip install "syntharc[all]"
+
+# Or individually pick your domains:
+pip install "syntharc[tabular]"
+pip install "syntharc[text]"
+pip install "syntharc[image]"
+pip install "syntharc[timeseries]"
+
+# For contributors and local development:
+pip install "syntharc[dev]"
+```
+
+---
+
+## 💡 Comprehensive Quick Start
+
+Below are examples of how our unified framework elegantly handles vastly different data constraints.
+
+### 1. Privacy-Preserving Tabular Data
+Train a CTGAN model to learn the distribution of your customer data without saving real identities.
+```python
+import pandas as pd
+from syntharc.tabular.ctgan import CTGANSynthesizer
+
+real_data = pd.read_csv("customer_data.csv")
+
+# 1. Initialize & Prepare metadata
+synth = CTGANSynthesizer(epochs=50)
+synth.prepare(metadata_dict={"primary_key": "user_id"})
+
+# 2. Fit to real data
+synth.fit(real_data)
+
+# 3. Generate 1,000 synthetic rows!
+synthetic_data = synth.generate(num_rows=1000)
+print(synthetic_data.head())
+```
+
+### 2. Deep Time-Series Generation
+Synthesize sequential metrics, like stock prices or IoT sensor readings, using AutoRegressive modeling.
+```python
+from syntharc.timeseries.par import PARSynthesizer
+
+# 1. Initialize & Prepare
+synth = PARSynthesizer(epochs=25)
+# Define what makes a "sequence" (e.g., separate tracking per 'device_id')
+synth.prepare(metadata_dict={
+    "sequence_key": "device_id",
+    "context_columns": ["region"]
+})
+
+# 2. Fit
+synth.fit(sensor_dataframe)
+
+# 3. Generate sequential data
+synthetic_series = synth.generate(num_sequences=50)
+```
+
+### 3. Causal Text Generation (LLMs)
+Easily utilize Hugging Face causal language models.
+```python
+from syntharc.text.transformer import TransformerTextGenerator
+
+# 1. Initialize (Downloads SmolLM2-360M-Instruct by default)
+synth = TransformerTextGenerator()
+
+# 2. Prepare with an optional style context
+synth.prepare(corpus="The future of synthetic data is bright and highly scalable.")
+
+# 3. Generate structured text from instructions
+text_output = synth.generate(num_samples=1, instructions="Write a short summary about data.")
+print(text_output[0])
+```
+
+### 4. High-Speed Image Augmentation
+Prepare a pipeline of augmentations to expand your Machine Learning dataset effortlessly.
+```python
+from syntharc.image.augmentor import ImageAugmentor
+import cv2
+
+# 1. Initialize rules
+augmentor = ImageAugmentor()
+# Uses Albumentations backend dictionary standards mapping
+augmentor.prepare(config={
+    'resize': (256, 256),
+    'horizontal_flip': 0.5,
+    'brightness_contrast': 0.2
+})
+
+# 2. Load Real Image
+image = cv2.imread('dataset/cat.jpg')
+
+# 3. Generate augmentations
+# The generator seamlessly wraps Albumentations under the hood
+augmented_image = augmentor.generate(num_samples=1, data=[image])
+print(f"Augmented Shape: {augmented_image[0].shape}")
+```
+
+---
+
+## 📖 Documentation
+
+For comprehensive guides, parameter references, evaluations, and interactive Jupyter showcase notebooks, please check out the `docs/` folder:
+- [Getting Started](docs/getting_started.md)
+- [Tabular Generation Guide](docs/tabular.md)
+- [Time-Series Generation Guide](docs/timeseries.md)
+- [Text Generation Guide](docs/text.md)
+- [Image Augmentation Guide](docs/image.md)
+- [API Reference](docs/api_reference.md)
+
+---
+
+## 🤝 Contributing
+
+We love our contributors! If you're interested in adding a new generative model, fixing a bug, or improving the documentation, please refer to our [CONTRIBUTING.md](CONTRIBUTING.md) for detailed instructions on setting up your local environment, managing dependencies, and passing our CI checks.
+
+## 📄 License
+
+This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
+
+---
+
+Made with ❤️ by the syntharc team

syntharc-0.1.0/pyproject.toml

@@ -0,0 +1,122 @@
+[build-system]
+requires = ["setuptools>=69.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "syntharc"
+version = "0.1.0"
+description = "Unified synthetic data generation for tabular, time-series, image, and text data"
+readme = "README.md"
+license = {file = "LICENSE"}
+requires-python = ">=3.10"
+authors = [
+    {name = "Fahad Abdullah", email = "fahadai.co@gmail.com"},
+]
+keywords = [
+    "synthetic-data",
+    "syntharceration",
+    "augmentation",
+    "tabular",
+    "time-series",
+    "image",
+    "text",
+]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Developers",
+    "Intended Audience :: Science/Research",
+    "License :: OSI Approved :: MIT License",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Topic :: Scientific/Engineering :: Artificial Intelligence",
+    "Topic :: Software Development :: Libraries :: Python Modules",
+]
+
+dependencies = [
+    "pandas>=1.5",
+    "numpy>=1.23",
+    "scikit-learn>=1.2",
+    "matplotlib>=3.6",
+    "Pillow>=9.0",
+    "pyyaml>=6.0",
+    "rich>=13.0",
+    "torch>=2.0",
+]
+
+[project.optional-dependencies]
+tabular = [
+    "sdv>=1.10",
+]
+timeseries = [
+    "sdv>=1.10",
+]
+image = [
+    "albumentations>=1.4",
+    "opencv-python-headless>=4.9",
+]
+text = [
+    "transformers>=4.40",
+    "accelerate>=0.22",
+]
+dev = [
+    "pytest>=8.0",
+    "pytest-cov>=5.0",
+    "ruff>=0.4",
+    "mypy>=1.10",
+    "pre-commit>=3.7",
+    "pandas-stubs",
+    "types-PyYAML",
+    "types-Pillow",
+]
+all = [
+    "syntharc[tabular]",
+    "syntharc[timeseries]",
+    "syntharc[image]",
+    "syntharc[text]",
+    "syntharc[dev]",
+]
+
+[project.urls]
+Homepage = "https://github.com/FAbdullah17/syntharc"
+Repository = "https://github.com/FAbdullah17/syntharc"
+Issues = "https://github.com/FAbdullah17/syntharc/issues"
+Documentation = "https://github.com/FAbdullah17/syntharc/tree/main/docs"
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[tool.ruff]
+target-version = "py310"
+line-length = 100
+src = ["src", "tests"]
+
+[tool.ruff.lint]
+select = [
+    "E",    # pycodestyle errors
+    "F",    # pyflakes
+    "W",    # pycodestyle warnings
+    "I",    # isort
+    "N",    # pep8-naming
+    "UP",   # pyupgrade
+    "B",    # flake8-bugbear
+    "SIM",  # flake8-simplify
+    "RUF",  # ruff-specific rules
+]
+
+[tool.ruff.lint.isort]
+known-first-party = ["syntharc"]
+
+[tool.mypy]
+python_version = "3.10"
+warn_return_any = true
+warn_unused_configs = true
+disallow_untyped_defs = true
+mypy_path = "src"
+packages = ["syntharc"]
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+addopts = "-v --tb=short"

syntharc-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

syntharc-0.1.0/src/syntharc/__init__.py

@@ -0,0 +1,38 @@
+"""syntharc — Unified synthetic data generation.
+
+A lightweight Python package for synthetic data generation across
+tabular, time-series, image, and text domains using sample-based
+learning, augmentation, and lightweight generative techniques.
+
+Quick Start
+-----------
+>>> from syntharc.core import BaseSynthesizer, set_seed, setup_logging
+
+Tabular (requires ``pip install syntharc[tabular]``):
+
+>>> from syntharc.tabular import CTGANSynthesizer  # doctest: +SKIP
+>>> from syntharc.tabular import GaussianCopulaSynthesizer  # doctest: +SKIP
+
+Time-series (requires ``pip install syntharc[timeseries]``):
+
+>>> from syntharc.timeseries import TimeSeriesSynthesizer  # doctest: +SKIP
+
+Image (requires ``pip install syntharc[image]``):
+
+>>> from syntharc.image import ImageAugmentor  # doctest: +SKIP
+
+Text (markov/template work out of the box, transformer needs
+``pip install syntharc[text]``):
+
+>>> from syntharc.text import MarkovTextGenerator  # doctest: +SKIP
+>>> from syntharc.text import TemplateTextGenerator  # doctest: +SKIP
+>>> from syntharc.text import TransformerTextGenerator  # doctest: +SKIP
+"""
+
+from __future__ import annotations
+
+__version__ = "0.1.0"
+__author__ = "Fahad Abdullah"
+__email__ = "fahadai.co@gmail.com"
+
+__all__ = ["__version__"]

syntharc-0.1.0/src/syntharc/core/__init__.py

@@ -0,0 +1,14 @@
+"""syntharc.core — Core infrastructure for syntharc."""
+
+from syntharc.core.base import BaseSynthesizer
+from syntharc.core.config import load_config, validate_config
+from syntharc.core.utils import get_device, set_seed, setup_logging
+
+__all__ = [
+    "BaseSynthesizer",
+    "load_config",
+    "validate_config",
+    "get_device",
+    "set_seed",
+    "setup_logging",
+]