FedModelKit 0.6.7__tar.gz → 0.7.0__tar.gz

fedmodelkit-0.7.0/.python-version

@@ -0,0 +1 @@
+3.12.7

fedmodelkit-0.7.0/PKG-INFO

@@ -0,0 +1,93 @@
+Metadata-Version: 2.4
+Name: FedModelKit
+Version: 0.7.0
+Summary: Tools to scaffold and run privacy-preserving federated learning experiments across distributed data sites.
+Author-email: ceresale <alessandro.ceresi@upm.es>
+License-File: LICENSE
+Requires-Python: >=3.12
+Requires-Dist: flwr-datasets>=0.5.0
+Requires-Dist: flwr[simulation]>=1.17.0
+Requires-Dist: mlflow>=3.1.1
+Requires-Dist: pandas>=2.3.0
+Provides-Extra: edge
+Description-Content-Type: text/markdown
+
+# 🌐 FedModelKit (fmk)
+
+**FedModelKit** is a Python library designed to streamline the execution of **federated learning tasks** within a network of collaborating partners.  
+It provides a guided workflow that helps users quickly set up experiments, manage dependencies, and follow step-by-step instructions through generated resources.
+
+---
+
+## 📑 Table of Contents
+- [Overview](#-overview)
+- [Features](#-features)
+- [Dependencies](#-Dependencies)
+- [Installation](#-installation)
+- [Quick Start](#-quick-start)
+- [Next Steps](#-next-steps)
+
+---
+
+## 🔎 Overview
+
+Federated learning enables multiple partners to train machine learning models collaboratively **without sharing raw data**.  
+The **FedModelKit** library simplifies this process by:
+
+- Automating project initialization  
+- Creating a reproducible experiment directory with all required dependencies
+- Offering a clear, step-by-step workflow to run experiments in a federated network    
+
+---
+
+## 🧩 Dependencies
+
+These are the dependencies of the packages, including the required Python version:
+
+| Dependency        | Version     | Description                                      |
+|-------------------|------------|--------------------------------------------------|
+| **Python**        | >=3.13        | Required Python interpreter version              |
+| `flwr[simulations]` | ==1.17.0  | Federated learning framework and simulations with Flower |
+| `flwr_datasets`   | >=0.5.0     | Datasets compatible with Flower                  |
+| `mlflow`          | >=3.1.1     | For experiment tracking and logging              |
+| `pandas`          | >=2.3.0     | Data manipulation and analysis                   |
+
+
+---
+
+## ⚙️ Installation
+
+It is recommended to install `FedModelKit` inside a virtual environment.  
+We suggest using [uv](https://github.com/astral-sh/uv) for fast and reproducible setups:
+
+
+1. Create and activate a virtual environment
+    ```bash
+    uv venv
+    source .venv/bin/activate
+    ```
+    or in case of Windows OS
+    ```bash
+    uv venv
+    .venv\Scripts\activate
+    ```
+
+2. Install `FedModelKit`
+    ```bash
+    uv pip install FedModelKit
+
+    ```
+
+## 🚀 Quick Start
+
+Initialize a new federated learning experiment with:
+
+```bash
+fmk init -n my_experiment
+
+```
+This will generate in the present directory the project for a new FL experiment.
+
+## 📖 Next Steps
+
+👉 Once the experiment directory has been created, open the file `README.md` that has been generated in your directory and follow the guided workflow to set up and run your federated learning tasks.

fedmodelkit-0.7.0/README.md

@@ -0,0 +1,79 @@
+# 🌐 FedModelKit (fmk)
+
+**FedModelKit** is a Python library designed to streamline the execution of **federated learning tasks** within a network of collaborating partners.  
+It provides a guided workflow that helps users quickly set up experiments, manage dependencies, and follow step-by-step instructions through generated resources.
+
+---
+
+## 📑 Table of Contents
+- [Overview](#-overview)
+- [Features](#-features)
+- [Dependencies](#-Dependencies)
+- [Installation](#-installation)
+- [Quick Start](#-quick-start)
+- [Next Steps](#-next-steps)
+
+---
+
+## 🔎 Overview
+
+Federated learning enables multiple partners to train machine learning models collaboratively **without sharing raw data**.  
+The **FedModelKit** library simplifies this process by:
+
+- Automating project initialization  
+- Creating a reproducible experiment directory with all required dependencies
+- Offering a clear, step-by-step workflow to run experiments in a federated network    
+
+---
+
+## 🧩 Dependencies
+
+These are the dependencies of the packages, including the required Python version:
+
+| Dependency        | Version     | Description                                      |
+|-------------------|------------|--------------------------------------------------|
+| **Python**        | >=3.13        | Required Python interpreter version              |
+| `flwr[simulations]` | ==1.17.0  | Federated learning framework and simulations with Flower |
+| `flwr_datasets`   | >=0.5.0     | Datasets compatible with Flower                  |
+| `mlflow`          | >=3.1.1     | For experiment tracking and logging              |
+| `pandas`          | >=2.3.0     | Data manipulation and analysis                   |
+
+
+---
+
+## ⚙️ Installation
+
+It is recommended to install `FedModelKit` inside a virtual environment.  
+We suggest using [uv](https://github.com/astral-sh/uv) for fast and reproducible setups:
+
+
+1. Create and activate a virtual environment
+    ```bash
+    uv venv
+    source .venv/bin/activate
+    ```
+    or in case of Windows OS
+    ```bash
+    uv venv
+    .venv\Scripts\activate
+    ```
+
+2. Install `FedModelKit`
+    ```bash
+    uv pip install FedModelKit
+
+    ```
+
+## 🚀 Quick Start
+
+Initialize a new federated learning experiment with:
+
+```bash
+fmk init -n my_experiment
+
+```
+This will generate in the present directory the project for a new FL experiment.
+
+## 📖 Next Steps
+
+👉 Once the experiment directory has been created, open the file `README.md` that has been generated in your directory and follow the guided workflow to set up and run your federated learning tasks.

fedmodelkit-0.7.0/pyproject.toml

@@ -0,0 +1,26 @@
+[project]
+name = "FedModelKit"
+version = "0.7.0"
+description = "Tools to scaffold and run privacy-preserving federated learning experiments across distributed data sites."
+readme = "README.md"
+authors = [
+    { name = "ceresale", email = "alessandro.ceresi@upm.es" }
+]
+requires-python = ">=3.12"
+dependencies = [
+    "flwr-datasets>=0.5.0",
+    "flwr[simulation]>=1.17.0",
+    "pandas>=2.3.0",
+    "mlflow>=3.1.1",
+]
+
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project.scripts]
+fmk = "FedModelKit.cli:main"
+
+[project.optional-dependencies]
+edge = [
+]

fedmodelkit-0.7.0/src/FedModelKit/README.md

@@ -0,0 +1,19 @@
+### **Interface Directory**
+
+1. **`aggregator.py`**
+    - This file defines the `AggProtocol` protocols. The `AggProtocol` defines the methods and attributes that an aggregator must implement.
+
+2. **`local_learner.py`**
+    - This file defines the `LLProtocol` protocols. The `LLProtocol` defines the methods and attributes that a local learner must implement. This file provides the structure for implementing custom local learners.
+
+3. **`cli.py`**
+    - This file contains the `create_structure` function, which sets up the project structure for launching a FL task. It creates necessary directories and files for the workflow.
+
+4. **`__init__.py`**
+    - This script exposes the main components of the FedModelKit package for easy import.
+
+5. **`templates/`**
+    - This folder contains template scripts for generating the content of the generated directory.
+
+6. **`mlflow_service/`**
+    - This folder contains scripts that define services for model and training registry in a Mlflow server.

fedmodelkit-0.7.0/src/FedModelKit/__init__.py

@@ -0,0 +1,10 @@
+'''from .interface import FederatedModel, submit_fl_model'''
+from .local_learner import LLProtocol
+from .aggregator import AggProtocol
+from .cli import create_structure
+import flwr
+
+__all__ = ["LLProtocol",
+           "AggProtocol",
+           "create_structure",
+           "flwr"]

{fedmodelkit-0.6.7 → fedmodelkit-0.7.0}/src/FedModelKit/aggregator.py

@@ -29,13 +29,3 @@ class AggProtocol(Protocol):
         """
         ...
 
-@runtime_checkable
-class AggFactoryProtocol(Protocol):
-    '''
-    Protocol for the function creating the aggregator instance
-    '''
-    def __call__(self) -> AggProtocol:
-        """
-        Creates and returns an instance of a class conforming to AggProtocol.
-        """
-        ...

{fedmodelkit-0.6.7 → fedmodelkit-0.7.0}/src/FedModelKit/cli.py

@@ -2,16 +2,16 @@ import os
 import sys
 import argparse
 from .templates import (
-    get_main_template,
-    get_task_template,
-    get_readme_template,
-    get_server_template,
-    get_client_template,
-    get_pyproject_template,
-    get_extern_pyproject_template,
-    get_uv_template,
-    get_init_template,
-    get_ds_template,
+    get_task,
+    get_readme,
+    get_server,
+    get_client,
+    get_pyproject,
+    get_extern_pyproject,
+    get_init,
+    get_ds,
+    get_do1,
+    get_do2,
     get_images,
 )
 
@@ -20,19 +20,19 @@ def create_structure(exp_name: str = "new_experiment") -> None:
 
     structure = {
         f"{exp_name}": {
-            "main.py": get_main_template(),
-            "pyproject.toml": get_pyproject_template(exp_name),
+            "pyproject.toml": get_pyproject(exp_name),
             f"{exp_name}": {
-                "__init__.py": get_init_template(),
-                "task.py": get_task_template(),
-                "server_app.py": get_server_template(exp_name),
-                "client_app.py": get_client_template(exp_name),                
+                "__init__.py": get_init(),
+                "task.py": get_task(),
+                "server_app.py": get_server(exp_name),
+                "client_app.py": get_client(exp_name),                
             },
         },
-        "pyproject.toml": get_extern_pyproject_template(exp_name),
-        "uv.lock": get_uv_template(),
-        "README.md": get_readme_template(),
-        "ds.ipynb": get_ds_template(exp_name),
+        "pyproject.toml": get_extern_pyproject(exp_name),
+        "README.md": get_readme(),
+        "ds.ipynb": get_ds(exp_name),
+        "do1.ipynb": get_do1(exp_name),
+        "do2.ipynb": get_do2(exp_name),
     }
 
 

{fedmodelkit-0.6.7 → fedmodelkit-0.7.0}/src/FedModelKit/local_learner.py

@@ -49,13 +49,3 @@ class LLProtocol(Protocol):
         """
         ...
 
-@runtime_checkable
-class LLFactoryProtocol(Protocol):
-    '''
-    Protocol for the function creating the local learner instance
-    '''
-    def __call__(self) -> LLProtocol:
-            """
-            Creates and returns an instance of a class conforming to LLProtocol.
-            """
-            ...

fedmodelkit-0.7.0/src/FedModelKit/mlflow_service/__init__.py

@@ -0,0 +1,33 @@
+"""
+MLflow Coordination Utilities for Federated Learning
+
+This package provides clean, consolidated MLflow coordination utilities for federated learning
+with Flower, enabling distributed logging, coordination, and monitoring that can
+be easily integrated into any federated learning project.
+"""
+
+from .messages import MLflowMessageType, MLflowCoordinationConfig, MLflowMetricsData, MLflowParamsData, MLflowRoundCompleteData, create_mlflow_coordination_message, extract_mlflow_coordination_config
+from .server import MLflowServerCoordinator, create_coordinator_from_environment
+from .client import MLflowClientLogger, MLflowClientIntegration, create_mlflow_client_integration, handle_mlflow_coordination_query, track_in_mlflow
+
+__all__ = [
+    # Message types and data structures
+    "MLflowMessageType",
+    "MLflowCoordinationConfig", 
+    "MLflowMetricsData",
+    "MLflowParamsData",
+    "MLflowRoundCompleteData",
+    "create_mlflow_coordination_message",
+    "extract_mlflow_coordination_config",
+    
+    # Server-side coordination
+    "MLflowServerCoordinator",
+    "create_coordinator_from_environment",
+    
+    # Client-side logging
+    "MLflowClientLogger",
+    "MLflowClientIntegration",
+    "create_mlflow_client_integration",
+    "handle_mlflow_coordination_query",
+    "track_in_mlflow",
+]