# 🌟 KerasFactory - Reusable Model Architecture Bricks in Keras 🌟
<div align="center">
<img src="docs/logo.png" width="350" alt="KerasFactory Logo"/>
<p><strong>Provided and maintained by <a href="https://unicolab.ai">🦄 UnicoLab</a></strong></p>
</div>
[](https://www.python.org/downloads/)
[](https://keras.io/)
[](https://opensource.org/licenses/MIT)
[](https://github.com/psf/black)
[](https://unicolab.ai)
[](https://unicolab.github.io/KerasFactory/)
**KerasFactory** is a comprehensive collection of reusable Keras layers and models specifically designed for tabular data processing, feature engineering, and advanced neural network architectures. Built with Keras 3 and developed by [🦄 UnicoLab](https://unicolab.ai), it provides a clean, efficient, and extensible foundation for building sophisticated machine learning models for enterprise AI applications.
## ✨ Key Features
- **🎯 38+ Production-Ready Layers**: Attention mechanisms, feature processing, preprocessing, and specialized architectures
- **🧠 Advanced Models**: SFNE blocks, Terminator models, and more coming soon
- **📊 Data Analyzer**: Intelligent CSV analysis tool that recommends appropriate layers
- **🔬 Experimental Modules**: 20+ cutting-edge layers and models for research
- **⚡ Keras 3 Only**: Pure Keras 3 implementation with no TensorFlow dependencies
- **🧪 Comprehensive Testing**: Full test coverage with 38+ test suites
- **📚 Rich Documentation**: Detailed guides, examples, and API documentation
## 🚀 Quick Start
### Installation
```bash
# Install from PyPI
poetry add kerasfactory
# Or install from source
git clone https://github.com/UnicoLab/KerasFactory
cd KerasFactory
poetry install
```
### 🚀 Quick Start Examples
#### Example 1: Smart Data Preprocessing
```python
import keras
from kerasfactory.layers import DistributionTransformLayer
# Create a simple model with automatic data transformation
inputs = keras.Input(shape=(10,)) # 10 numerical features
# Automatically transform data to normal distribution
transformed = DistributionTransformLayer(transform_type='auto')(inputs)
# Simple neural network
x = keras.layers.Dense(64, activation='relu')(transformed)
x = keras.layers.Dropout(0.2)(x)
outputs = keras.layers.Dense(1, activation='sigmoid')(x)
model = keras.Model(inputs=inputs, outputs=outputs)
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
print("Model ready! The layer will automatically choose the best transformation for your data.")
```
#### Example 2: Intelligent Feature Fusion
```python
import keras
from kerasfactory.layers import GatedFeatureFusion
# Create two different representations of your data
inputs = keras.Input(shape=(8,)) # 8 features
# First representation: linear processing
linear_features = keras.layers.Dense(16, activation='relu')(inputs)
# Second representation: non-linear processing
nonlinear_features = keras.layers.Dense(16, activation='tanh')(inputs)
# Intelligently combine both representations
fused_features = GatedFeatureFusion()([linear_features, nonlinear_features])
# Final prediction
outputs = keras.layers.Dense(1, activation='sigmoid')(fused_features)
model = keras.Model(inputs=inputs, outputs=outputs)
model.compile(optimizer='adam', loss='binary_crossentropy')
print("Smart feature fusion model ready! The layer learns which representation to trust more.")
```
#### Example 3: Ready-to-Use Models
```python
import keras
from kerasfactory.models import BaseFeedForwardModel
# Create a complete model with just one line!
model = BaseFeedForwardModel(
feature_names=['age', 'income', 'education', 'experience'],
hidden_units=[64, 32, 16],
output_units=1,
dropout_rate=0.2
)
# Your data (each feature as separate input)
age = keras.random.normal((100, 1))
income = keras.random.normal((100, 1))
education = keras.random.normal((100, 1))
experience = keras.random.normal((100, 1))
# Train with one command
model.compile(optimizer='adam', loss='mse')
model.fit([age, income, education, experience],
keras.random.normal((100, 1)),
epochs=10, verbose=0)
print("✅ Model trained successfully! No complex setup needed.")
```
#### Example 4: Date Feature Engineering
```python
import keras
from kerasfactory.layers import DateEncodingLayer
# Create a model that processes date information
inputs = keras.Input(shape=(4,)) # [year, month, day, day_of_week]
# Convert dates to cyclical features automatically
date_features = DateEncodingLayer()(inputs)
# Simple prediction model
x = keras.layers.Dense(32, activation='relu')(date_features)
outputs = keras.layers.Dense(1, activation='sigmoid')(x)
model = keras.Model(inputs=inputs, outputs=outputs)
model.compile(optimizer='adam', loss='binary_crossentropy')
print("📅 Date-aware model ready! Handles seasonality and cyclical patterns automatically.")
```
### 🧠 Smart Data Analyzer
```python
from kerasfactory.utils import analyze_data
# Get intelligent recommendations for your data
results = analyze_data("your_data.csv")
recommendations = results["recommendations"]
print("🎯 Recommended layers for your data:")
for layer in recommendations:
print(f" • {layer['layer_name']}: {layer['description']}")
print("✨ No more guessing which layers to use!")
```
## 🏗️ Architecture Overview
### Core Components
#### **Layers** (`kerasfactory.layers`)
- **Attention Mechanisms**: `TabularAttention`, `MultiResolutionTabularAttention`, `ColumnAttention`, `RowAttention`
- **Feature Processing**: `AdvancedNumericalEmbedding`, `GatedFeatureFusion`, `VariableSelection`
- **Preprocessing**: `DateEncodingLayer`, `DateParsingLayer`, `DifferentiableTabularPreprocessor`
- **Advanced Architectures**: `TransformerBlock`, `GatedResidualNetwork`, `BoostingBlock`
- **Specialized Layers**: `BusinessRulesLayer`, `StochasticDepth`, `FeatureCutout`
#### **Models** (`kerasfactory.models`)
- **SFNEBlock**: Advanced feature processing block
- **TerminatorModel**: Multi-block hierarchical processing model
#### **Utilities** (`kerasfactory.utils`)
- **Data Analyzer**: Intelligent CSV analysis and layer recommendation system
- **CLI Tools**: Command-line interface for data analysis
#### **Experimental** (`experimental/`)
- **Time Series**: 12+ specialized time series preprocessing layers
- **Advanced Models**: Neural Additive Models, Temporal Fusion Transformers, and more
- **Research Components**: Cutting-edge architectures for experimentation
- **Note**: Experimental components are not included in the PyPI package
## 📖 Documentation
- **[Online Documentation](https://unicolab.github.io/KerasFactory/)**: Full API reference with automatic docstring generation
- **[API Reference](https://unicolab.github.io/KerasFactory/api/)**: Complete documentation for all layers, models, and utilities
- **[Layer Implementation Guide](docs/layers_implementation_guide.md)**: Comprehensive guide for implementing new layers
- **[Data Analyzer Documentation](docs/data_analyzer.md)**: Complete guide to the data analysis tools
- **[Contributing Guide](docs/contributing.md)**: How to contribute to the project
## 🎯 Common Use Cases
### 📊 Tabular Data Processing
```python
from kerasfactory.layers import DistributionTransformLayer, GatedFeatureFusion
# Smart preprocessing
preprocessor = DistributionTransformLayer(transform_type='auto')
# Feature combination
fusion = GatedFeatureFusion()
```
### 🔧 Feature Engineering
```python
from kerasfactory.layers import DateEncodingLayer, BusinessRulesLayer
# Date features
date_encoder = DateEncodingLayer()
# Business rules validation
rules = BusinessRulesLayer(
rules=[(">", 0), ("<", 100)],
feature_type="numerical"
)
```
### 🎨 Advanced Architectures
```python
from kerasfactory.layers import StochasticDepth, GatedResidualNetwork
# Regularization
stochastic_depth = StochasticDepth(survival_prob=0.8)
# Advanced processing
grn = GatedResidualNetwork(units=64)
```
## 🧪 Testing
```bash
# Run all tests
make all_tests
# Run specific test categories
make unittests
make data_analyzer_tests
# Generate coverage report
make coverage
```
## 🤝 Contributing
We welcome contributions! Please see our [Contributing Guide](docs/contributing.md) for details.
### Development Setup
```bash
# Clone the repository
git clone https://github.com/UnicoLab/KerasFactory.git
cd KerasFactory
# Install development dependencies
poetry install
# Install pre-commit hooks
pre-commit install
# Run tests
make all_tests
```
### Commit Convention
We use semantic commit messages:
- `feat(KerasFactory): add new layer for feature processing`
- `fix(KerasFactory): resolve serialization issue`
- `docs(KerasFactory): update installation guide`
## 📊 Performance
KerasFactory is optimized for performance with:
- **Keras 3 Backend**: Leverages the latest Keras optimizations
- **Efficient Operations**: Uses only Keras operations for maximum compatibility
- **Memory Optimization**: Careful memory management in complex layers
- **Batch Processing**: Optimized for batch operations
## 💬 Join Our Community
Have questions or want to connect with other KDP users? Join us on Discord:
[](https://discord.gg/6zf4VZFYV5)
## 📄 License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## 🙏 Acknowledgments
- Built with [Keras 3](https://keras.io/)
- Inspired by modern deep learning research
- Community-driven development
## 📞 Support
- **Issues**: [GitHub Issues](https://github.com/UnicoLab/KerasFactory/issues)
- **Discussions**: [GitHub Discussions](https://github.com/UnicoLab/KerasFactory/discussions)
- **Documentation**: [Online Docs](https://unicolab.github.io/KerasFactory/)
- **Discord**: [](https://discord.gg/6zf4VZFYV5)
---
<p align="center">
<strong>Built with ❤️ for the Keras community by 🦄 UnicoLab.ai</strong>
</p>
Raw data
{
"_id": null,
"home_page": "https://unicolab.ai",
"name": "kerasfactory",
"maintainer": null,
"docs_url": null,
"requires_python": "<3.13,>=3.10",
"maintainer_email": null,
"keywords": "keras, tensorflow, machine-learning, deep-learning, tabular-data, attention, neural-networks, unicolab, enterprise-ai",
"author": "UnicoLab",
"author_email": "contact@unicolab.ai",
"download_url": "https://files.pythonhosted.org/packages/11/7a/8794595f6f183079ac5b39e27b4438e16c01daa5fd9edf64ec23c81bbab0/kerasfactory-0.1.0.tar.gz",
"platform": null,
"description": "# \ud83c\udf1f KerasFactory - Reusable Model Architecture Bricks in Keras \ud83c\udf1f\n\n<div align=\"center\">\n <img src=\"docs/logo.png\" width=\"350\" alt=\"KerasFactory Logo\"/>\n \n <p><strong>Provided and maintained by <a href=\"https://unicolab.ai\">\ud83e\udd84 UnicoLab</a></strong></p>\n</div>\n\n[](https://www.python.org/downloads/)\n[](https://keras.io/)\n[](https://opensource.org/licenses/MIT)\n[](https://github.com/psf/black)\n[](https://unicolab.ai)\n[](https://unicolab.github.io/KerasFactory/)\n\n**KerasFactory** is a comprehensive collection of reusable Keras layers and models specifically designed for tabular data processing, feature engineering, and advanced neural network architectures. Built with Keras 3 and developed by [\ud83e\udd84 UnicoLab](https://unicolab.ai), it provides a clean, efficient, and extensible foundation for building sophisticated machine learning models for enterprise AI applications.\n\n## \u2728 Key Features\n\n- **\ud83c\udfaf 38+ Production-Ready Layers**: Attention mechanisms, feature processing, preprocessing, and specialized architectures\n- **\ud83e\udde0 Advanced Models**: SFNE blocks, Terminator models, and more coming soon\n- **\ud83d\udcca Data Analyzer**: Intelligent CSV analysis tool that recommends appropriate layers\n- **\ud83d\udd2c Experimental Modules**: 20+ cutting-edge layers and models for research\n- **\u26a1 Keras 3 Only**: Pure Keras 3 implementation with no TensorFlow dependencies\n- **\ud83e\uddea Comprehensive Testing**: Full test coverage with 38+ test suites\n- **\ud83d\udcda Rich Documentation**: Detailed guides, examples, and API documentation\n\n## \ud83d\ude80 Quick Start\n\n### Installation\n\n```bash\n# Install from PyPI\npoetry add kerasfactory\n\n# Or install from source\ngit clone https://github.com/UnicoLab/KerasFactory\ncd KerasFactory\npoetry install\n```\n\n### \ud83d\ude80 Quick Start Examples\n\n#### Example 1: Smart Data Preprocessing\n\n```python\nimport keras\nfrom kerasfactory.layers import DistributionTransformLayer\n\n# Create a simple model with automatic data transformation\ninputs = keras.Input(shape=(10,)) # 10 numerical features\n\n# Automatically transform data to normal distribution\ntransformed = DistributionTransformLayer(transform_type='auto')(inputs)\n\n# Simple neural network\nx = keras.layers.Dense(64, activation='relu')(transformed)\nx = keras.layers.Dropout(0.2)(x)\noutputs = keras.layers.Dense(1, activation='sigmoid')(x)\n\nmodel = keras.Model(inputs=inputs, outputs=outputs)\nmodel.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n\nprint(\"Model ready! The layer will automatically choose the best transformation for your data.\")\n```\n\n#### Example 2: Intelligent Feature Fusion\n\n```python\nimport keras\nfrom kerasfactory.layers import GatedFeatureFusion\n\n# Create two different representations of your data\ninputs = keras.Input(shape=(8,)) # 8 features\n\n# First representation: linear processing\nlinear_features = keras.layers.Dense(16, activation='relu')(inputs)\n\n# Second representation: non-linear processing \nnonlinear_features = keras.layers.Dense(16, activation='tanh')(inputs)\n\n# Intelligently combine both representations\nfused_features = GatedFeatureFusion()([linear_features, nonlinear_features])\n\n# Final prediction\noutputs = keras.layers.Dense(1, activation='sigmoid')(fused_features)\n\nmodel = keras.Model(inputs=inputs, outputs=outputs)\nmodel.compile(optimizer='adam', loss='binary_crossentropy')\n\nprint(\"Smart feature fusion model ready! The layer learns which representation to trust more.\")\n```\n\n#### Example 3: Ready-to-Use Models\n\n```python\nimport keras\nfrom kerasfactory.models import BaseFeedForwardModel\n\n# Create a complete model with just one line!\nmodel = BaseFeedForwardModel(\n feature_names=['age', 'income', 'education', 'experience'],\n hidden_units=[64, 32, 16],\n output_units=1,\n dropout_rate=0.2\n)\n\n# Your data (each feature as separate input)\nage = keras.random.normal((100, 1))\nincome = keras.random.normal((100, 1)) \neducation = keras.random.normal((100, 1))\nexperience = keras.random.normal((100, 1))\n\n# Train with one command\nmodel.compile(optimizer='adam', loss='mse')\nmodel.fit([age, income, education, experience], \n keras.random.normal((100, 1)), \n epochs=10, verbose=0)\n\nprint(\"\u2705 Model trained successfully! No complex setup needed.\")\n```\n\n#### Example 4: Date Feature Engineering\n\n```python\nimport keras\nfrom kerasfactory.layers import DateEncodingLayer\n\n# Create a model that processes date information\ninputs = keras.Input(shape=(4,)) # [year, month, day, day_of_week]\n\n# Convert dates to cyclical features automatically\ndate_features = DateEncodingLayer()(inputs)\n\n# Simple prediction model\nx = keras.layers.Dense(32, activation='relu')(date_features)\noutputs = keras.layers.Dense(1, activation='sigmoid')(x)\n\nmodel = keras.Model(inputs=inputs, outputs=outputs)\nmodel.compile(optimizer='adam', loss='binary_crossentropy')\n\nprint(\"\ud83d\udcc5 Date-aware model ready! Handles seasonality and cyclical patterns automatically.\")\n```\n\n### \ud83e\udde0 Smart Data Analyzer\n\n```python\nfrom kerasfactory.utils import analyze_data\n\n# Get intelligent recommendations for your data\nresults = analyze_data(\"your_data.csv\")\nrecommendations = results[\"recommendations\"]\n\nprint(\"\ud83c\udfaf Recommended layers for your data:\")\nfor layer in recommendations:\n print(f\" \u2022 {layer['layer_name']}: {layer['description']}\")\n \nprint(\"\u2728 No more guessing which layers to use!\")\n```\n\n## \ud83c\udfd7\ufe0f Architecture Overview\n\n### Core Components\n\n#### **Layers** (`kerasfactory.layers`)\n- **Attention Mechanisms**: `TabularAttention`, `MultiResolutionTabularAttention`, `ColumnAttention`, `RowAttention`\n- **Feature Processing**: `AdvancedNumericalEmbedding`, `GatedFeatureFusion`, `VariableSelection`\n- **Preprocessing**: `DateEncodingLayer`, `DateParsingLayer`, `DifferentiableTabularPreprocessor`\n- **Advanced Architectures**: `TransformerBlock`, `GatedResidualNetwork`, `BoostingBlock`\n- **Specialized Layers**: `BusinessRulesLayer`, `StochasticDepth`, `FeatureCutout`\n\n#### **Models** (`kerasfactory.models`)\n- **SFNEBlock**: Advanced feature processing block\n- **TerminatorModel**: Multi-block hierarchical processing model\n\n#### **Utilities** (`kerasfactory.utils`)\n- **Data Analyzer**: Intelligent CSV analysis and layer recommendation system\n- **CLI Tools**: Command-line interface for data analysis\n\n#### **Experimental** (`experimental/`)\n- **Time Series**: 12+ specialized time series preprocessing layers\n- **Advanced Models**: Neural Additive Models, Temporal Fusion Transformers, and more\n- **Research Components**: Cutting-edge architectures for experimentation\n- **Note**: Experimental components are not included in the PyPI package\n\n## \ud83d\udcd6 Documentation\n\n- **[Online Documentation](https://unicolab.github.io/KerasFactory/)**: Full API reference with automatic docstring generation\n- **[API Reference](https://unicolab.github.io/KerasFactory/api/)**: Complete documentation for all layers, models, and utilities\n- **[Layer Implementation Guide](docs/layers_implementation_guide.md)**: Comprehensive guide for implementing new layers\n- **[Data Analyzer Documentation](docs/data_analyzer.md)**: Complete guide to the data analysis tools\n- **[Contributing Guide](docs/contributing.md)**: How to contribute to the project\n\n## \ud83c\udfaf Common Use Cases\n\n### \ud83d\udcca Tabular Data Processing\n```python\nfrom kerasfactory.layers import DistributionTransformLayer, GatedFeatureFusion\n\n# Smart preprocessing\npreprocessor = DistributionTransformLayer(transform_type='auto')\n\n# Feature combination\nfusion = GatedFeatureFusion()\n```\n\n### \ud83d\udd27 Feature Engineering\n```python\nfrom kerasfactory.layers import DateEncodingLayer, BusinessRulesLayer\n\n# Date features\ndate_encoder = DateEncodingLayer()\n\n# Business rules validation\nrules = BusinessRulesLayer(\n rules=[(\">\", 0), (\"<\", 100)], \n feature_type=\"numerical\"\n)\n```\n\n### \ud83c\udfa8 Advanced Architectures\n```python\nfrom kerasfactory.layers import StochasticDepth, GatedResidualNetwork\n\n# Regularization\nstochastic_depth = StochasticDepth(survival_prob=0.8)\n\n# Advanced processing\ngrn = GatedResidualNetwork(units=64)\n```\n\n## \ud83e\uddea Testing\n\n```bash\n# Run all tests\nmake all_tests\n\n# Run specific test categories\nmake unittests\nmake data_analyzer_tests\n\n# Generate coverage report\nmake coverage\n```\n\n## \ud83e\udd1d Contributing\n\nWe welcome contributions! Please see our [Contributing Guide](docs/contributing.md) for details.\n\n### Development Setup\n\n```bash\n# Clone the repository\ngit clone https://github.com/UnicoLab/KerasFactory.git\ncd KerasFactory\n\n# Install development dependencies\npoetry install\n\n# Install pre-commit hooks\npre-commit install\n\n# Run tests\nmake all_tests\n```\n\n### Commit Convention\n\nWe use semantic commit messages:\n- `feat(KerasFactory): add new layer for feature processing`\n- `fix(KerasFactory): resolve serialization issue`\n- `docs(KerasFactory): update installation guide`\n\n## \ud83d\udcca Performance\n\nKerasFactory is optimized for performance with:\n- **Keras 3 Backend**: Leverages the latest Keras optimizations\n- **Efficient Operations**: Uses only Keras operations for maximum compatibility\n- **Memory Optimization**: Careful memory management in complex layers\n- **Batch Processing**: Optimized for batch operations\n\n## \ud83d\udcac Join Our Community\n\nHave questions or want to connect with other KDP users? Join us on Discord:\n\n[](https://discord.gg/6zf4VZFYV5)\n\n## \ud83d\udcc4 License\n\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\n\n## \ud83d\ude4f Acknowledgments\n\n- Built with [Keras 3](https://keras.io/)\n- Inspired by modern deep learning research\n- Community-driven development\n\n## \ud83d\udcde Support\n\n- **Issues**: [GitHub Issues](https://github.com/UnicoLab/KerasFactory/issues)\n- **Discussions**: [GitHub Discussions](https://github.com/UnicoLab/KerasFactory/discussions)\n- **Documentation**: [Online Docs](https://unicolab.github.io/KerasFactory/)\n- **Discord**: [](https://discord.gg/6zf4VZFYV5)\n\n---\n\n<p align=\"center\">\n <strong>Built with \u2764\ufe0f for the Keras community by \ud83e\udd84 UnicoLab.ai</strong>\n</p>",
"bugtrack_url": null,
"license": "MIT",
"summary": "Reusable Model Architecture Bricks in Keras - Enterprise AI by UnicoLab",
"version": "0.1.0",
"project_urls": {
"Documentation": "https://unicolab.github.io/KerasFactory/",
"Homepage": "https://unicolab.ai",
"Repository": "https://github.com/UnicoLab/KerasFactory"
},
"split_keywords": [
"keras",
" tensorflow",
" machine-learning",
" deep-learning",
" tabular-data",
" attention",
" neural-networks",
" unicolab",
" enterprise-ai"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "5b42ebd8b5ceccdc6e99df035891ea3522eddaca9acc1c5e413de4baa583254d",
"md5": "5b7ee2fb5bbab1486aa859bc73b5dfbf",
"sha256": "c5b8db31d095b517816132190388a98ffb050d3cf8cae6e37001e6fc7c32d12b"
},
"downloads": -1,
"filename": "kerasfactory-0.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "5b7ee2fb5bbab1486aa859bc73b5dfbf",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<3.13,>=3.10",
"size": 175153,
"upload_time": "2025-11-13T10:17:48",
"upload_time_iso_8601": "2025-11-13T10:17:48.534616Z",
"url": "https://files.pythonhosted.org/packages/5b/42/ebd8b5ceccdc6e99df035891ea3522eddaca9acc1c5e413de4baa583254d/kerasfactory-0.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "117a8794595f6f183079ac5b39e27b4438e16c01daa5fd9edf64ec23c81bbab0",
"md5": "85f0aef0241278f5ce727ee8ef6cecb1",
"sha256": "42ebf6afb40d553b882802a41e546131cf032e8f8c0314408b3e6738fd914a7b"
},
"downloads": -1,
"filename": "kerasfactory-0.1.0.tar.gz",
"has_sig": false,
"md5_digest": "85f0aef0241278f5ce727ee8ef6cecb1",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<3.13,>=3.10",
"size": 115854,
"upload_time": "2025-11-13T10:17:50",
"upload_time_iso_8601": "2025-11-13T10:17:50.158832Z",
"url": "https://files.pythonhosted.org/packages/11/7a/8794595f6f183079ac5b39e27b4438e16c01daa5fd9edf64ec23c81bbab0/kerasfactory-0.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-11-13 10:17:50",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "UnicoLab",
"github_project": "KerasFactory",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "kerasfactory"
}
JSON
