outdatedLabs 0.1.0__tar.gz → 0.1.1__tar.gz

{outdatedlabs-0.1.0 → outdatedlabs-0.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: outdatedLabs
-Version: 0.1.0
+Version: 0.1.1
 Summary: A secure package for training machine learning models
 Home-page: https://github.com/Mouli51ch/OutDated
 Author: OutDated Team
@@ -30,9 +30,9 @@ Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary
 
-# OutDatedLabs - Secure Machine Learning Training Package
+# OutdatedLabs - Secure Machine Learning Training Package
 
-A Python package for secure machine learning model training using the ML training server.
+A secure and easy-to-use package for training machine learning models with remote server support.
 
 ## Installation
 
@@ -42,73 +42,65 @@ pip install outdatedLabs
 
 ## Quick Start
 
+1. Start the ML training server:
+```bash
+cd mlTrainingServer
+uv run main.py
+```
+
+2. Use the package in your code:
 ```python
 from outdatedLabs import SecureModel
 
 # Create a linear regression model
 model = SecureModel.linearRegression()
 
-# Optionally set a custom server URL
-model.set_server_url("http://your-server:8000")
-
 # Train the model
-model.fit(
+model.train(
     dataset_hash="your_dataset_hash",
     features=["feature1", "feature2"],
     target="target_column"
 )
 
-# Make predictions
-predictions = model.predict(X)
-
 # Get training metrics
 metrics = model.get_metrics()
 print(metrics)
+
+# Download and load the trained model
+model.download_model()
+loaded_model = model.load_model()
 ```
 
 ## Features
 
-- Secure model training using encrypted datasets
-- Support for linear regression models
+- Secure model training with remote server support
+- Automatic dataset download and cleanup
 - Progress tracking with tqdm
-- Comprehensive error handling and logging
-- Easy-to-use interface
-- Configurable server URL
+- Comprehensive error handling
+- Detailed logging
+- Support for multiple algorithms (currently Linear Regression)
+
+## Configuration
+
+The package connects to a local ML training server by default at `http://localhost:3000`. You can change the server URL when creating a model:
+
+```python
+model = SecureModel.linearRegression(server_url="http://your-server:3000")
+```
 
 ## API Reference
 
 ### SecureModel
 
-The main class for secure model training.
+The main class for model training and management.
 
 #### Methods
 
-- `linearRegression(server_url: str = "http://localhost:8000") -> SecureModel`
-  - Create a new linear regression model instance
-  
-- `set_server_url(server_url: str) -> SecureModel`
-  - Set a custom URL for the ML training server
-  
-- `fit(dataset_hash: str, features: List[str] = None, target: str = None, params: Dict[str, Any] = None) -> SecureModel`
-  - Train the model using the specified dataset
-  
-- `predict(X: Union[pd.DataFrame, List[List[float]]]) -> List[float]`
-  - Make predictions using the trained model
-  
-- `score(X: Union[pd.DataFrame, List[List[float]]], y: List[float]) -> Dict[str, float]`
-  - Calculate model performance metrics
-  
-- `get_metrics() -> Dict[str, Any]`
-  - Get training metrics
-
-## Requirements
-
-- Python 3.7+
-- requests
-- joblib
-- pandas
-- scikit-learn
-- tqdm
+- `linearRegression(server_url: str = "http://localhost:3000")`: Create a Linear Regression model
+- `train(dataset_hash: str, features: List[str], target: str)`: Train the model
+- `get_metrics() -> Dict`: Get training metrics
+- `download_model() -> str`: Download the trained model
+- `load_model() -> Any`: Load the trained model
 
 ## License
 

outdatedlabs-0.1.1/README.md

@@ -0,0 +1,75 @@
+# OutdatedLabs - Secure Machine Learning Training Package
+
+A secure and easy-to-use package for training machine learning models with remote server support.
+
+## Installation
+
+```bash
+pip install outdatedLabs
+```
+
+## Quick Start
+
+1. Start the ML training server:
+```bash
+cd mlTrainingServer
+uv run main.py
+```
+
+2. Use the package in your code:
+```python
+from outdatedLabs import SecureModel
+
+# Create a linear regression model
+model = SecureModel.linearRegression()
+
+# Train the model
+model.train(
+    dataset_hash="your_dataset_hash",
+    features=["feature1", "feature2"],
+    target="target_column"
+)
+
+# Get training metrics
+metrics = model.get_metrics()
+print(metrics)
+
+# Download and load the trained model
+model.download_model()
+loaded_model = model.load_model()
+```
+
+## Features
+
+- Secure model training with remote server support
+- Automatic dataset download and cleanup
+- Progress tracking with tqdm
+- Comprehensive error handling
+- Detailed logging
+- Support for multiple algorithms (currently Linear Regression)
+
+## Configuration
+
+The package connects to a local ML training server by default at `http://localhost:3000`. You can change the server URL when creating a model:
+
+```python
+model = SecureModel.linearRegression(server_url="http://your-server:3000")
+```
+
+## API Reference
+
+### SecureModel
+
+The main class for model training and management.
+
+#### Methods
+
+- `linearRegression(server_url: str = "http://localhost:3000")`: Create a Linear Regression model
+- `train(dataset_hash: str, features: List[str], target: str)`: Train the model
+- `get_metrics() -> Dict`: Get training metrics
+- `download_model() -> str`: Download the trained model
+- `load_model() -> Any`: Load the trained model
+
+## License
+
+MIT License 

{outdatedlabs-0.1.0 → outdatedlabs-0.1.1}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup, find_packages
 
 setup(
     name="outdatedLabs",
-    version="0.1.0",
+    version="0.1.1",
     packages=find_packages(where="src"),
     package_dir={"": "src"},
     install_requires=[

{outdatedlabs-0.1.0 → outdatedlabs-0.1.1}/src/outdatedLabs/secure_model.py

@@ -23,7 +23,7 @@ logger = logging.getLogger(__name__)
 class SecureModel:
     """Secure Model Training Class"""
     
-    def __init__(self, server_url: str = "http://localhost:8000"):
+    def __init__(self, server_url: str = "http://localhost:3000"):
         """
         Initialize the SecureModel.
         
@@ -35,6 +35,18 @@ class SecureModel:
         self.model = None
         self.metrics = None
         
+        # Test server connection
+        try:
+            response = requests.get(f"{self.server_url}/health")
+            response.raise_for_status()
+            logger.info(f"Successfully connected to server at {self.server_url}")
+        except requests.exceptions.ConnectionError:
+            logger.error(f"Could not connect to server at {self.server_url}. Please ensure the server is running.")
+            raise
+        except requests.exceptions.RequestException as e:
+            logger.error(f"Error connecting to server: {str(e)}")
+            raise
+        
     def set_server_url(self, server_url: str) -> 'SecureModel':
         """
         Set the training server URL.
@@ -47,10 +59,23 @@ class SecureModel:
         """
         self.server_url = server_url.rstrip('/')
         logger.info(f"Server URL set to: {self.server_url}")
+        
+        # Test new server connection
+        try:
+            response = requests.get(f"{self.server_url}/health")
+            response.raise_for_status()
+            logger.info(f"Successfully connected to new server at {self.server_url}")
+        except requests.exceptions.ConnectionError:
+            logger.error(f"Could not connect to server at {self.server_url}. Please ensure the server is running.")
+            raise
+        except requests.exceptions.RequestException as e:
+            logger.error(f"Error connecting to server: {str(e)}")
+            raise
+            
         return self
         
     @classmethod
-    def linearRegression(cls, server_url: str = "http://localhost:8000") -> 'SecureModel':
+    def linearRegression(cls, server_url: str = "http://localhost:3000") -> 'SecureModel':
         """
         Create a Linear Regression model.
         
@@ -124,20 +149,29 @@ class SecureModel:
         """Wait for the training job to complete."""
         with tqdm(desc="Training", unit="s") as pbar:
             while True:
-                response = requests.get(f"{self.server_url}/train/{self.job_id}/status")
-                response.raise_for_status()
-                status = response.json()
-                
-                if status["status"] == "COMPLETE":
-                    logger.info("Training completed successfully")
-                    self.metrics = requests.get(f"{self.server_url}/train/{self.job_id}/metrics").json()
-                    break
-                elif status["status"] == "FAILED":
-                    error = status.get("error", "Unknown error")
-                    raise Exception(f"Training failed: {error}")
-                
-                time.sleep(check_interval)
-                pbar.update(check_interval)
+                try:
+                    response = requests.get(f"{self.server_url}/train/{self.job_id}/status")
+                    response.raise_for_status()
+                    status = response.json()
+                    
+                    # Convert status to uppercase for case-insensitive comparison
+                    current_status = status["status"].upper()
+                    
+                    if current_status == "COMPLETE":
+                        logger.info("Training completed successfully")
+                        self.metrics = requests.get(f"{self.server_url}/train/{self.job_id}/metrics").json()
+                        pbar.close()
+                        return
+                    elif current_status == "FAILED":
+                        error = status.get("error", "Unknown error")
+                        pbar.close()
+                        raise Exception(f"Training failed: {error}")
+                    
+                    time.sleep(check_interval)
+                    pbar.update(check_interval)
+                except Exception as e:
+                    pbar.close()
+                    raise Exception(f"Error checking training status: {str(e)}")
     
     def _download_model(self) -> None:
         """Download the trained model."""

{outdatedlabs-0.1.0 → outdatedlabs-0.1.1}/src/outdatedLabs.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: outdatedLabs
-Version: 0.1.0
+Version: 0.1.1
 Summary: A secure package for training machine learning models
 Home-page: https://github.com/Mouli51ch/OutDated
 Author: OutDated Team
@@ -30,9 +30,9 @@ Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary
 
-# OutDatedLabs - Secure Machine Learning Training Package
+# OutdatedLabs - Secure Machine Learning Training Package
 
-A Python package for secure machine learning model training using the ML training server.
+A secure and easy-to-use package for training machine learning models with remote server support.
 
 ## Installation
 
@@ -42,73 +42,65 @@ pip install outdatedLabs
 
 ## Quick Start
 
+1. Start the ML training server:
+```bash
+cd mlTrainingServer
+uv run main.py
+```
+
+2. Use the package in your code:
 ```python
 from outdatedLabs import SecureModel
 
 # Create a linear regression model
 model = SecureModel.linearRegression()
 
-# Optionally set a custom server URL
-model.set_server_url("http://your-server:8000")
-
 # Train the model
-model.fit(
+model.train(
     dataset_hash="your_dataset_hash",
     features=["feature1", "feature2"],
     target="target_column"
 )
 
-# Make predictions
-predictions = model.predict(X)
-
 # Get training metrics
 metrics = model.get_metrics()
 print(metrics)
+
+# Download and load the trained model
+model.download_model()
+loaded_model = model.load_model()
 ```
 
 ## Features
 
-- Secure model training using encrypted datasets
-- Support for linear regression models
+- Secure model training with remote server support
+- Automatic dataset download and cleanup
 - Progress tracking with tqdm
-- Comprehensive error handling and logging
-- Easy-to-use interface
-- Configurable server URL
+- Comprehensive error handling
+- Detailed logging
+- Support for multiple algorithms (currently Linear Regression)
+
+## Configuration
+
+The package connects to a local ML training server by default at `http://localhost:3000`. You can change the server URL when creating a model:
+
+```python
+model = SecureModel.linearRegression(server_url="http://your-server:3000")
+```
 
 ## API Reference
 
 ### SecureModel
 
-The main class for secure model training.
+The main class for model training and management.
 
 #### Methods
 
-- `linearRegression(server_url: str = "http://localhost:8000") -> SecureModel`
-  - Create a new linear regression model instance
-  
-- `set_server_url(server_url: str) -> SecureModel`
-  - Set a custom URL for the ML training server
-  
-- `fit(dataset_hash: str, features: List[str] = None, target: str = None, params: Dict[str, Any] = None) -> SecureModel`
-  - Train the model using the specified dataset
-  
-- `predict(X: Union[pd.DataFrame, List[List[float]]]) -> List[float]`
-  - Make predictions using the trained model
-  
-- `score(X: Union[pd.DataFrame, List[List[float]]], y: List[float]) -> Dict[str, float]`
-  - Calculate model performance metrics
-  
-- `get_metrics() -> Dict[str, Any]`
-  - Get training metrics
-
-## Requirements
-
-- Python 3.7+
-- requests
-- joblib
-- pandas
-- scikit-learn
-- tqdm
+- `linearRegression(server_url: str = "http://localhost:3000")`: Create a Linear Regression model
+- `train(dataset_hash: str, features: List[str], target: str)`: Train the model
+- `get_metrics() -> Dict`: Get training metrics
+- `download_model() -> str`: Download the trained model
+- `load_model() -> Any`: Load the trained model
 
 ## License
 

outdatedlabs-0.1.0/README.md

@@ -1,83 +0,0 @@
-# OutDatedLabs - Secure Machine Learning Training Package
-
-A Python package for secure machine learning model training using the ML training server.
-
-## Installation
-
-```bash
-pip install outdatedLabs
-```
-
-## Quick Start
-
-```python
-from outdatedLabs import SecureModel
-
-# Create a linear regression model
-model = SecureModel.linearRegression()
-
-# Optionally set a custom server URL
-model.set_server_url("http://your-server:8000")
-
-# Train the model
-model.fit(
-    dataset_hash="your_dataset_hash",
-    features=["feature1", "feature2"],
-    target="target_column"
-)
-
-# Make predictions
-predictions = model.predict(X)
-
-# Get training metrics
-metrics = model.get_metrics()
-print(metrics)
-```
-
-## Features
-
-- Secure model training using encrypted datasets
-- Support for linear regression models
-- Progress tracking with tqdm
-- Comprehensive error handling and logging
-- Easy-to-use interface
-- Configurable server URL
-
-## API Reference
-
-### SecureModel
-
-The main class for secure model training.
-
-#### Methods
-
-- `linearRegression(server_url: str = "http://localhost:8000") -> SecureModel`
-  - Create a new linear regression model instance
-  
-- `set_server_url(server_url: str) -> SecureModel`
-  - Set a custom URL for the ML training server
-  
-- `fit(dataset_hash: str, features: List[str] = None, target: str = None, params: Dict[str, Any] = None) -> SecureModel`
-  - Train the model using the specified dataset
-  
-- `predict(X: Union[pd.DataFrame, List[List[float]]]) -> List[float]`
-  - Make predictions using the trained model
-  
-- `score(X: Union[pd.DataFrame, List[List[float]]], y: List[float]) -> Dict[str, float]`
-  - Calculate model performance metrics
-  
-- `get_metrics() -> Dict[str, Any]`
-  - Get training metrics
-
-## Requirements
-
-- Python 3.7+
-- requests
-- joblib
-- pandas
-- scikit-learn
-- tqdm
-
-## License
-
-MIT License