{"id":14970751,"url":"https://github.com/piotrlaczkowski/keras-data-processor","last_synced_at":"2025-05-06T21:00:25.989Z","repository":{"id":227667416,"uuid":"769308510","full_name":"piotrlaczkowski/keras-data-processor","owner":"piotrlaczkowski","description":"Data Preprocessing model based on Keras preprocessing layers that can be used as a standalone model or incorporated to Keras model as first layers.","archived":false,"fork":false,"pushed_at":"2025-04-11T11:30:49.000Z","size":11279,"stargazers_count":5,"open_issues_count":4,"forks_count":5,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-04-19T14:58:17.363Z","etag":null,"topics":["data","keras","layers","preprocessing","tensorflow"],"latest_commit_sha":null,"homepage":"https://piotrlaczkowski.github.io/keras-data-processor/latest","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/piotrlaczkowski.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"docs/contributing/development/auto-documentation.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":".github/CODEOWNERS","security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2024-03-08T19:22:05.000Z","updated_at":"2025-04-17T09:28:39.000Z","dependencies_parsed_at":"2025-04-19T09:32:49.487Z","dependency_job_id":"e3d87742-4777-4a54-9c87-dc1e38e86033","html_url":"https://github.com/piotrlaczkowski/keras-data-processor","commit_stats":null,"previous_names":["piotrlaczkowski/keras-data-processor"],"tags_count":12,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/piotrlaczkowski%2Fkeras-data-processor","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/piotrlaczkowski%2Fkeras-data-processor/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/piotrlaczkowski%2Fkeras-data-processor/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/piotrlaczkowski%2Fkeras-data-processor/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/piotrlaczkowski","download_url":"https://codeload.github.com/piotrlaczkowski/keras-data-processor/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":252769396,"owners_count":21801375,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["data","keras","layers","preprocessing","tensorflow"],"created_at":"2024-09-24T13:44:05.485Z","updated_at":"2025-05-06T21:00:25.976Z","avatar_url":"https://github.com/piotrlaczkowski.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🌟 Keras Data Processor (KDP) - Powerful Data Preprocessing for TensorFlow\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"docs/getting-started/imgs/kdp_logo.png\" width=\"350\"/\u003e\n\u003c/p\u003e\n\n**Transform your raw data into ML-ready features with just a few lines of code!**\n\nKDP provides a state-of-the-art preprocessing system built on TensorFlow Keras. It handles everything from feature normalization to advanced embedding techniques, making your ML pipelines faster, more robust, and easier to maintain.\n\n## ✨ Key Features\n\n- 🚀 **Efficient Single-Pass Processing**: Process all features in one go, dramatically faster than alternatives\n- 🧠 **Distribution-Aware Encoding**: Automatically detects and optimally handles different data distributions\n- 👁️ **Tabular Attention**: Captures complex feature interactions for better model performance\n- 🔍 **Feature Selection**: Automatically identifies and focuses on the most important features\n- 🔄 **Feature-wise Mixture of Experts**: Specialized processing for different feature types\n- 📦 **Production-Ready**: Deploy your preprocessing along with your model as a single unit\n\n## 🚀 Quick Installation\n\n```bash\n# Using pip\npip install keras-data-processor\n\n# Using Poetry\npoetry add keras-data-processor\n```\n\n## 📋 Simple Example\n\n```python\nfrom kdp import PreprocessingModel, FeatureType\n\n# Define your features\nfeatures_specs = {\n    \"age\": FeatureType.FLOAT_NORMALIZED,\n    \"income\": FeatureType.FLOAT_RESCALED,\n    \"occupation\": FeatureType.STRING_CATEGORICAL,\n    \"description\": FeatureType.TEXT\n}\n\n# Create and build the preprocessor\npreprocessor = PreprocessingModel(\n    path_data=\"data/my_data.csv\",\n    features_specs=features_specs,\n    # Enable advanced features\n    use_distribution_aware=True,\n    tabular_attention=True\n)\nresult = preprocessor.build_preprocessor()\nmodel = result[\"model\"]\n\n# Use the preprocessor with your data\nprocessed_features = model(input_data)\n```\n\n## 📚 Comprehensive Documentation\n\nWe've built an extensive documentation system to help you get the most from KDP:\n\n### Core Guides\n\n- [🚀 Quick Start Guide](docs/quick_start.md) - Get up and running in minutes\n- [📊 Feature Processing](docs/features.md) - Learn about all supported feature types\n- [🧙‍♂️ Auto-Configuration](docs/auto_configuration.md) - Let KDP configure itself for your data\n\n### Advanced Topics\n\n- [📈 Distribution-Aware Encoding](docs/distribution_aware_encoder.md) - Smart handling of different distributions\n- [👁️ Tabular Attention](docs/tabular_attention.md) - Capture complex feature interactions\n- [🔢 Advanced Numerical Embeddings](docs/advanced_numerical_embeddings.md) - Rich representations for numbers\n- [🤖 Transformer Blocks](docs/transformer_blocks.md) - Apply transformer architecture to tabular data\n- [🎯 Feature Selection](docs/feature_selection.md) - Focus on what matters in your data\n- [🧠 Feature-wise Mixture of Experts](docs/feature_moe.md) - Specialized processing per feature\n\n### Integration \u0026 Performance\n\n- [🔗 Integration Guide](docs/integrations.md) - Use KDP with existing ML pipelines\n- [🚀 Tabular Optimization](docs/tabular_optimization.md) - Supercharge your preprocessing\n- [📈 Performance Tips](docs/complex_examples.md) - Handling large datasets efficiently\n\n### Background \u0026 Resources\n\n- [💡 Motivation](docs/motivation.md) - Why we built KDP\n- [🤝 Contributing](docs/contributing.md) - Help improve KDP\n\n## 🖼️ Model Architecture\n\nYour preprocessing pipeline is built as a Keras model that can be used independently or as the first layer of any model:\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"docs/features/imgs/models/all_basic_types.png\" width=\"800\"/\u003e\n\u003c/p\u003e\n\n## 📊 Performance\n\nKDP outperforms alternative preprocessing approaches, especially as data size increases:\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"docs/getting-started/imgs/time_vs_nr_data.png\" width=\"400\"/\u003e\n  \u003cimg src=\"docs/getting-started/imgs/time_vs_nr_features.png\" width=\"400\"/\u003e\n\u003c/p\u003e\n\n## 🤝 Contributing\n\nWe welcome contributions! Please check out our [Contributing Guide](docs/contributing.md) for guidelines on how to proceed.\n\n## 🛠️ Development Tools\n\nKDP includes tools to help developers:\n\n- **Documentation Generation**: Automatically generate API docs from docstrings\n- **Model Diagram Generation**: Visualize model architectures with `make generate_doc_content` or run:\n  ```bash\n  python scripts/generate_model_diagrams.py\n  ```\n  This creates diagram images in `docs/features/imgs/models/` for all feature types and configurations.\n\n## 📄 License\n\nThis project is licensed under the MIT License - see the LICENSE file for details.\n\n## 🙏 Acknowledgments\n\n- The TensorFlow and Keras teams for their amazing work\n- All contributors who help make KDP better\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpiotrlaczkowski%2Fkeras-data-processor","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpiotrlaczkowski%2Fkeras-data-processor","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpiotrlaczkowski%2Fkeras-data-processor/lists"}