collie-mlops 0.1.1b0__tar.gz → 0.1.2b0__tar.gz

{collie-mlops-0.1.1b0 → collie_mlops-0.1.2b0}/MANIFEST.in

@@ -9,6 +9,7 @@ recursive-include collie py.typed
 recursive-exclude tests *
 recursive-exclude example *
 recursive-exclude deploy *
+recursive-exclude collie/helper *
 recursive-exclude __pycache__ *
 recursive-exclude *.pyc
 recursive-exclude *.pyo

{collie-mlops-0.1.1b0 → collie_mlops-0.1.2b0}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: collie-mlops
-Version: 0.1.1b0
+Version: 0.1.2b0
 Summary: A Lightweight MLOps Framework for Machine Learning Workflows
 Home-page: https://github.com/ChingHuanChiu/collie
 Author: ChingHuanChiu
@@ -29,17 +29,46 @@ Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Classifier: Typing :: Typed
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: mlflow>=2.0.0
+Requires-Dist: pydantic>=2.0.0
+Requires-Dist: pandas>=1.3.0
+Requires-Dist: numpy<2.0.0,>=1.20.0
 Provides-Extra: sklearn
+Requires-Dist: scikit-learn>=1.0.0; extra == "sklearn"
 Provides-Extra: xgboost
-Provides-Extra: pytorch
+Requires-Dist: xgboost>=1.5.0; extra == "xgboost"
 Provides-Extra: lightgbm
-Provides-Extra: transformers
-Provides-Extra: config
-Provides-Extra: dev
-Provides-Extra: test
-Provides-Extra: docs
+Requires-Dist: lightgbm>=3.0.0; extra == "lightgbm"
+Provides-Extra: pytorch
+Requires-Dist: torch>=1.9.0; extra == "pytorch"
+Requires-Dist: pytorch-lightning>=2.0.0; extra == "pytorch"
+Requires-Dist: transformers>=4.0.0; extra == "pytorch"
+Requires-Dist: sentence-transformers>=2.0.0; extra == "pytorch"
+Provides-Extra: tabular
+Requires-Dist: scikit-learn>=1.0.0; extra == "tabular"
+Requires-Dist: xgboost>=1.5.0; extra == "tabular"
+Requires-Dist: lightgbm>=3.0.0; extra == "tabular"
+Provides-Extra: deep-learning
+Requires-Dist: torch>=1.9.0; extra == "deep-learning"
+Requires-Dist: pytorch-lightning>=2.0.0; extra == "deep-learning"
+Requires-Dist: transformers>=4.0.0; extra == "deep-learning"
+Requires-Dist: sentence-transformers>=2.0.0; extra == "deep-learning"
 Provides-Extra: all
-License-File: LICENSE
+Requires-Dist: scikit-learn>=1.0.0; extra == "all"
+Requires-Dist: xgboost>=1.5.0; extra == "all"
+Requires-Dist: lightgbm>=3.0.0; extra == "all"
+Requires-Dist: torch>=1.9.0; extra == "all"
+Requires-Dist: pytorch-lightning>=2.0.0; extra == "all"
+Requires-Dist: transformers>=4.0.0; extra == "all"
+Requires-Dist: sentence-transformers>=2.0.0; extra == "all"
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Requires-Dist: pytest-cov>=3.0.0; extra == "dev"
+Requires-Dist: black>=22.0.0; extra == "dev"
+Requires-Dist: mypy>=0.990; extra == "dev"
+Requires-Dist: ruff>=0.0.260; extra == "dev"
+Dynamic: license-file
 
 # Collie
 
@@ -79,16 +108,42 @@ Collie follows an event-driven architecture with the following core components:
 
 ### Installation
 
+#### Basic Installation (Core Framework Only)
 ```bash
 pip install collie-mlops
 ```
 
-This will install Collie with all supported ML frameworks including:
-- scikit-learn
-- PyTorch
-- XGBoost
-- LightGBM
-- Transformers (with Sentence Transformers)
+This installs the core MLOps orchestration framework with MLflow integration (~100MB).
+
+#### Install with ML Frameworks
+
+Choose the installation that fits your needs:
+
+**For Traditional ML (Tabular Data)**
+```bash
+# Individual frameworks
+pip install collie-mlops[sklearn]      # scikit-learn support
+pip install collie-mlops[xgboost]      # XGBoost support
+pip install collie-mlops[lightgbm]     # LightGBM support
+
+# Or install all tabular ML frameworks (~250MB)
+pip install collie-mlops[tabular]
+```
+
+**For Deep Learning**
+```bash
+# PyTorch ecosystem (includes Transformers for NLP/Vision) (~3GB)
+pip install collie-mlops[pytorch]
+
+# Or use the alias
+pip install collie-mlops[deep-learning]
+```
+
+**For Complete Installation**
+```bash
+# All frameworks (~3.5GB)
+pip install collie-mlops[all]
+```
 
 ### Prerequisites
 
@@ -157,6 +212,32 @@ class CustomPusher(Pusher):
         return Event(payload=PusherPayload(model_uri=model_uri))
 ```
 
+### Orchestrator
+Coordinates the execution of all components in the pipeline.
+
+```python
+from collie import Orchestrator
+
+# Create orchestrator with all components
+orchestrator = Orchestrator(
+    components=[
+        CustomTransformer(),
+        CustomTuner(),
+        CustomTrainer(),
+        CustomEvaluator(),
+        CustomPusher()
+    ],
+    tracking_uri="http://localhost:5000",
+    experiment_name="my_experiment",
+    registered_model_name="my_model",
+    mlflow_tags={"project": "my_project"},
+    description="My ML Pipeline"
+)
+
+# Run the entire pipeline
+orchestrator.run()
+```
+
 ## Configuration
 
 ### MLflow Setup
@@ -173,13 +254,27 @@ mlflow server \
 
 ## Supported Frameworks
 
-Collie supports multiple ML frameworks through its model flavor system currently:
+Collie supports multiple ML frameworks through its flexible optional dependency system:
+
+### Available Frameworks
+-  **scikit-learn** - Traditional ML algorithms
+-  **XGBoost** - Gradient boosting for tabular data
+-  **LightGBM** - Fast gradient boosting framework  
+-  **PyTorch** - Deep learning framework
+-  **PyTorch Lightning** - High-level PyTorch wrapper
+-  **Transformers** - Hugging Face transformers for NLP
+-  **Sentence Transformers** - Sentence embeddings
+
+### Installation Options
 
--  **PyTorch** 
--  **scikit-learn**
--  **XGBoost** 
--  **LightGBM**
--  **Transformers**
+| Use Case | Command | Size | Frameworks Included |
+|----------|---------|------|---------------------|
+| **Core Only** | `pip install collie-mlops` | ~100MB | MLflow orchestration only |
+| **Tabular ML** | `pip install collie-mlops[tabular]` | ~250MB | sklearn, XGBoost, LightGBM |
+| **Deep Learning** | `pip install collie-mlops[pytorch]` | ~3GB | PyTorch, Lightning, Transformers |
+| **Complete** | `pip install collie-mlops[all]` | ~3.5GB | All frameworks |
+
+> **Note**: Install only what you need to keep your environment lightweight!
 
 
 ## Documentation
@@ -188,9 +283,13 @@ Collie supports multiple ML frameworks through its model flavor system currently
 
 ## Roadmap
 
+### Core Features
+- [ ] **Pipeline Checkpoint & Resume** - Save intermediate results and resume from failure points
+
+### Framework Support
 - [ ] TensorFlow/Keras support
 - [ ] Model monitoring and drift detection
-- [ ] Integrate an LLM training/fine-tuning framework
+
 
 ## License
 

collie-mlops-0.1.1b0/collie_mlops.egg-info/PKG-INFO → collie_mlops-0.1.2b0/README.md

@@ -1,46 +1,3 @@
-Metadata-Version: 2.1
-Name: collie-mlops
-Version: 0.1.1b0
-Summary: A Lightweight MLOps Framework for Machine Learning Workflows
-Home-page: https://github.com/ChingHuanChiu/collie
-Author: ChingHuanChiu
-Author-email: ChingHuanChiu <stevenchiou8@gmail.com>
-Maintainer-email: ChingHuanChiu <stevenchiou8@gmail.com>
-License: MIT
-Project-URL: Homepage, https://github.com/ChingHuanChiu/collie
-Project-URL: Documentation, https://github.com/ChingHuanChiu/collie/blob/main/README.md
-Project-URL: Repository, https://github.com/ChingHuanChiu/collie
-Project-URL: Bug Tracker, https://github.com/ChingHuanChiu/collie/issues
-Project-URL: Changelog, https://github.com/ChingHuanChiu/collie/blob/main/CHANGELOG.md
-Keywords: mlops,machine-learning,mlflow,pipeline,orchestration,deep-learning,experiment-tracking
-Classifier: Development Status :: 4 - Beta
-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: Programming Language :: Python :: 3.13
-Classifier: Programming Language :: Python :: 3 :: Only
-Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Classifier: Topic :: Software Development :: Libraries :: Python Modules
-Classifier: Typing :: Typed
-Requires-Python: >=3.10
-Description-Content-Type: text/markdown
-Provides-Extra: sklearn
-Provides-Extra: xgboost
-Provides-Extra: pytorch
-Provides-Extra: lightgbm
-Provides-Extra: transformers
-Provides-Extra: config
-Provides-Extra: dev
-Provides-Extra: test
-Provides-Extra: docs
-Provides-Extra: all
-License-File: LICENSE
-
 # Collie
 
 [![PyPI version](https://badge.fury.io/py/collie-mlops.svg)](https://badge.fury.io/py/collie-mlops)
@@ -79,16 +36,42 @@ Collie follows an event-driven architecture with the following core components:
 
 ### Installation
 
+#### Basic Installation (Core Framework Only)
 ```bash
 pip install collie-mlops
 ```
 
-This will install Collie with all supported ML frameworks including:
-- scikit-learn
-- PyTorch
-- XGBoost
-- LightGBM
-- Transformers (with Sentence Transformers)
+This installs the core MLOps orchestration framework with MLflow integration (~100MB).
+
+#### Install with ML Frameworks
+
+Choose the installation that fits your needs:
+
+**For Traditional ML (Tabular Data)**
+```bash
+# Individual frameworks
+pip install collie-mlops[sklearn]      # scikit-learn support
+pip install collie-mlops[xgboost]      # XGBoost support
+pip install collie-mlops[lightgbm]     # LightGBM support
+
+# Or install all tabular ML frameworks (~250MB)
+pip install collie-mlops[tabular]
+```
+
+**For Deep Learning**
+```bash
+# PyTorch ecosystem (includes Transformers for NLP/Vision) (~3GB)
+pip install collie-mlops[pytorch]
+
+# Or use the alias
+pip install collie-mlops[deep-learning]
+```
+
+**For Complete Installation**
+```bash
+# All frameworks (~3.5GB)
+pip install collie-mlops[all]
+```
 
 ### Prerequisites
 
@@ -157,6 +140,32 @@ class CustomPusher(Pusher):
         return Event(payload=PusherPayload(model_uri=model_uri))
 ```
 
+### Orchestrator
+Coordinates the execution of all components in the pipeline.
+
+```python
+from collie import Orchestrator
+
+# Create orchestrator with all components
+orchestrator = Orchestrator(
+    components=[
+        CustomTransformer(),
+        CustomTuner(),
+        CustomTrainer(),
+        CustomEvaluator(),
+        CustomPusher()
+    ],
+    tracking_uri="http://localhost:5000",
+    experiment_name="my_experiment",
+    registered_model_name="my_model",
+    mlflow_tags={"project": "my_project"},
+    description="My ML Pipeline"
+)
+
+# Run the entire pipeline
+orchestrator.run()
+```
+
 ## Configuration
 
 ### MLflow Setup
@@ -173,13 +182,27 @@ mlflow server \
 
 ## Supported Frameworks
 
-Collie supports multiple ML frameworks through its model flavor system currently:
+Collie supports multiple ML frameworks through its flexible optional dependency system:
 
--  **PyTorch** 
--  **scikit-learn**
--  **XGBoost** 
--  **LightGBM**
--  **Transformers**
+### Available Frameworks
+-  **scikit-learn** - Traditional ML algorithms
+-  **XGBoost** - Gradient boosting for tabular data
+-  **LightGBM** - Fast gradient boosting framework  
+-  **PyTorch** - Deep learning framework
+-  **PyTorch Lightning** - High-level PyTorch wrapper
+-  **Transformers** - Hugging Face transformers for NLP
+-  **Sentence Transformers** - Sentence embeddings
+
+### Installation Options
+
+| Use Case | Command | Size | Frameworks Included |
+|----------|---------|------|---------------------|
+| **Core Only** | `pip install collie-mlops` | ~100MB | MLflow orchestration only |
+| **Tabular ML** | `pip install collie-mlops[tabular]` | ~250MB | sklearn, XGBoost, LightGBM |
+| **Deep Learning** | `pip install collie-mlops[pytorch]` | ~3GB | PyTorch, Lightning, Transformers |
+| **Complete** | `pip install collie-mlops[all]` | ~3.5GB | All frameworks |
+
+> **Note**: Install only what you need to keep your environment lightweight!
 
 
 ## Documentation
@@ -188,9 +211,13 @@ Collie supports multiple ML frameworks through its model flavor system currently
 
 ## Roadmap
 
+### Core Features
+- [ ] **Pipeline Checkpoint & Resume** - Save intermediate results and resume from failure points
+
+### Framework Support
 - [ ] TensorFlow/Keras support
 - [ ] Model monitoring and drift detection
-- [ ] Integrate an LLM training/fine-tuning framework
+
 
 ## License
 

{collie-mlops-0.1.1b0 → collie_mlops-0.1.2b0}/collie/contracts/mlflow.py

@@ -212,7 +212,7 @@ class _MLflowModelManager:
     def get_latest_version(
         self,
         model_name: str,
-        stages
+        stages: List[str]
     ) -> Optional[str]:
        
         """
@@ -275,6 +275,8 @@ class MLFlowComponentABC(ABC):
     def model_manager(self) -> Optional[_MLflowModelManager]:
 
         if self._model_manager is None:
+            if self._mlflow_config is None:
+                raise MLflowConfigurationError("MLflow config not set")
             mlflow_client = self._mlflow_config.mlflow_client
             self._model_manager = _MLflowModelManager(mlflow_client)
         return self._model_manager
@@ -310,10 +312,11 @@ class MLFlowComponentABC(ABC):
     
     def load_model(
         self, 
-        name: Optional[str] = None
+        flavor: ModelFlavor,
+        model_uri: Optional[str] = None
     ) -> Any:
        
-        return self.model_manager.load_model(name)
+        return self.model_manager.load_model(flavor, model_uri)
     
     def load_latest_model(
             self, 
@@ -355,7 +358,7 @@ class MLFlowComponentABC(ABC):
         self, 
         model_name: str, 
         stages: List[Literal["None", "Staging", "Production", "Archived"]]
-    ) -> int:
+    ) -> Optional[str]:
         
         return self.model_manager.get_latest_version(model_name, stages)
 
@@ -392,6 +395,9 @@ class MLFlowComponentABC(ABC):
             return None
         try:
             experiment = self.mlflow.get_experiment_by_name(experiment_name)
+            if experiment is None:
+                logger.warning(f"Experiment '{experiment_name}' not found")
+                return None
             if return_id:
                 return experiment.experiment_id
             else: