hqde 0.1.4__py3-none-any.whl → 0.1.5__py3-none-any.whl

hqde/core/hqde_system.py

@@ -152,27 +152,61 @@ class DistributedEnsembleManager:
             print(f"Running in simulated mode with {num_workers} workers (Ray not available)")
 
     def create_ensemble_workers(self, model_class, model_kwargs: Dict[str, Any]):
-        """Create distributed ensemble workers."""
+        """Create diverse distributed ensemble workers with different configurations."""
         # Calculate GPU fraction per worker (divide available GPUs among workers)
         num_gpus = torch.cuda.device_count() if torch.cuda.is_available() else 0
         gpu_per_worker = num_gpus / self.num_workers if num_gpus > 0 else 0
         
+        # ✅ FIX #6: ADD DIVERSITY - Different learning rates and dropout for each worker
+        learning_rates = [0.001, 0.0008, 0.0012, 0.0009][:self.num_workers]
+        dropout_rates = [0.15, 0.18, 0.12, 0.16][:self.num_workers]
+        
+        # Extend if more workers than predefined configs
+        while len(learning_rates) < self.num_workers:
+            learning_rates.append(0.001)
+            dropout_rates.append(0.15)
+        
         @ray.remote(num_gpus=gpu_per_worker)
         class EnsembleWorker:
-            def __init__(self, model_class, model_kwargs):
+            def __init__(self, model_class, model_kwargs, worker_id=0, learning_rate=0.001, dropout_rate=0.15):
+                # ✅ FIX #3: INJECT LOWER DROPOUT RATE
+                if 'dropout_rate' not in model_kwargs:
+                    model_kwargs['dropout_rate'] = dropout_rate
+                
                 self.model = model_class(**model_kwargs)
                 self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
                 self.model.to(self.device)
                 self.efficiency_score = 1.0
                 self.quantizer = AdaptiveQuantizer()
                 self.optimizer = None
+                self.scheduler = None
                 self.criterion = None
-
-            def setup_training(self, learning_rate=0.001):
-                """Setup optimizer and criterion for training."""
+                self.learning_rate = learning_rate
+                self.worker_id = worker_id
+
+            def setup_training(self, learning_rate=None):
+                """Setup optimizer, scheduler, and criterion for training."""
+                if learning_rate is None:
+                    learning_rate = self.learning_rate
+                
                 self.optimizer = torch.optim.Adam(self.model.parameters(), lr=learning_rate)
+                
+                # ✅ FIX #5: ADD LEARNING RATE SCHEDULING
+                self.scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(
+                    self.optimizer,
+                    T_max=50,  # Will be adjusted based on total epochs
+                    eta_min=1e-6
+                )
+                
                 self.criterion = torch.nn.CrossEntropyLoss()
                 return True
+            
+            def step_scheduler(self):
+                """Step the learning rate scheduler (call once per epoch)."""
+                if self.scheduler is not None:
+                    self.scheduler.step()
+                    return self.optimizer.param_groups[0]['lr']
+                return self.learning_rate
 
             def train_step(self, data_batch, targets=None):
                 # Perform actual training step using instance optimizer and criterion
@@ -187,6 +221,10 @@ class DistributedEnsembleManager:
                     outputs = self.model(data_batch)
                     loss = self.criterion(outputs, targets)
                     loss.backward()
+                    
+                    # ✅ GRADIENT CLIPPING for stability
+                    torch.nn.utils.clip_grad_norm_(self.model.parameters(), max_norm=1.0)
+                    
                     self.optimizer.step()
 
                     # Update efficiency score based on actual loss
@@ -222,8 +260,17 @@ class DistributedEnsembleManager:
                     outputs = self.model(data_batch)
                     return outputs.cpu()  # Move back to CPU for aggregation
 
-        self.workers = [EnsembleWorker.remote(model_class, model_kwargs)
-                       for _ in range(self.num_workers)]
+        # Create workers with diversity
+        self.workers = []
+        for worker_id in range(self.num_workers):
+            worker = EnsembleWorker.remote(
+                model_class, 
+                model_kwargs.copy(),  # Copy to avoid mutation
+                worker_id=worker_id,
+                learning_rate=learning_rates[worker_id],
+                dropout_rate=dropout_rates[worker_id]
+            )
+            self.workers.append(worker)
 
     def setup_workers_training(self, learning_rate=0.001):
         """Setup training for all workers."""
@@ -265,7 +312,7 @@ class DistributedEnsembleManager:
         ray.get(futures)
 
     def train_ensemble(self, data_loader, num_epochs: int = 10):
-        """Train the ensemble using distributed workers."""
+        """Train the ensemble using distributed workers with FedAvg-style aggregation."""
         # Setup training for all workers
         self.setup_workers_training()
 
@@ -298,15 +345,19 @@ class DistributedEnsembleManager:
                 batch_losses = ray.get(training_futures)
                 epoch_losses.extend([loss for loss in batch_losses if loss is not None])
 
-            # Only aggregate weights at the end of training (not after each epoch)
-            # This allows each worker to learn independently
-            # if epoch == num_epochs - 1:  # Only aggregate on last epoch
-            #     aggregated_weights = self.aggregate_weights()
-            #     if aggregated_weights:
-            #         self.broadcast_weights(aggregated_weights)
+            # ✅ FIX #1: AGGREGATE WEIGHTS AFTER EACH EPOCH (FedAvg style)
+            aggregated_weights = self.aggregate_weights()
+            if aggregated_weights:
+                self.broadcast_weights(aggregated_weights)
+                self.logger.info(f"  → Weights aggregated and synchronized at epoch {epoch + 1}")
+            
+            # ✅ FIX #5: STEP LEARNING RATE SCHEDULERS
+            scheduler_futures = [worker.step_scheduler.remote() for worker in self.workers]
+            current_lrs = ray.get(scheduler_futures)
+            avg_lr = np.mean(current_lrs) if current_lrs else 0.001
 
             avg_loss = np.mean(epoch_losses) if epoch_losses else 0.0
-            print(f"Epoch {epoch + 1}/{num_epochs}, Average Loss: {avg_loss:.4f}")
+            print(f"Epoch {epoch + 1}/{num_epochs}, Average Loss: {avg_loss:.4f}, LR: {avg_lr:.6f}")
 
     def shutdown(self):
         """Shutdown the distributed ensemble manager."""

hqde-0.1.5.dist-info/METADATA

@@ -0,0 +1,445 @@
+Metadata-Version: 2.4
+Name: hqde
+Version: 0.1.5
+Summary: Hierarchical Quantum-Distributed Ensemble Learning Framework
+Author-email: HQDE Team <hqde@example.com>
+Maintainer-email: HQDE Team <hqde@example.com>
+License: MIT
+Project-URL: Homepage, https://github.com/Prathmesh333/Hierarchical-Quantum-Distributed-Ensemble-Learning
+Project-URL: Repository, https://github.com/Prathmesh333/Hierarchical-Quantum-Distributed-Ensemble-Learning
+Project-URL: Documentation, https://github.com/Prathmesh333/Hierarchical-Quantum-Distributed-Ensemble-Learning/blob/main/HOW_TO_RUN.md
+Project-URL: Bug Reports, https://github.com/Prathmesh333/Hierarchical-Quantum-Distributed-Ensemble-Learning/issues
+Keywords: machine-learning,quantum,distributed,ensemble,deep-learning,pytorch,ray
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Science/Research
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy>=2.0.2
+Requires-Dist: torch>=2.8.0
+Requires-Dist: torchvision>=0.23.0
+Requires-Dist: ray[default]>=2.49.2
+Requires-Dist: scikit-learn>=1.6.1
+Requires-Dist: psutil>=7.1.0
+Provides-Extra: dev
+Requires-Dist: pytest>=8.4.2; extra == "dev"
+Requires-Dist: ipython>=8.18.1; extra == "dev"
+Requires-Dist: matplotlib>=3.9.4; extra == "dev"
+Requires-Dist: pandas>=2.3.2; extra == "dev"
+Requires-Dist: seaborn>=0.13.2; extra == "dev"
+Requires-Dist: tqdm>=4.67.1; extra == "dev"
+Dynamic: license-file
+
+# HQDE - Hierarchical Quantum-Distributed Ensemble Learning
+
+[![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/)
+[![PyTorch](https://img.shields.io/badge/PyTorch-2.8+-red.svg)](https://pytorch.org/)
+[![Ray](https://img.shields.io/badge/Ray-2.49+-green.svg)](https://ray.io/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+
+A production-ready framework for distributed ensemble learning with quantum-inspired algorithms and adaptive quantization.
+
+HQDE combines quantum-inspired algorithms with distributed computing to deliver superior machine learning performance with significantly reduced memory usage and training time.
+
+## Table of Contents
+
+- [Key Features](#key-features)
+- [Installation](#installation)
+- [Quick Start](#quick-start)
+- [Architecture Overview](#architecture-overview)
+- [Quantum-Inspired Algorithms](#quantum-inspired-algorithms)
+- [Distributed Computing](#distributed-computing)
+- [Adaptive Quantization](#adaptive-quantization)
+- [Configuration](#configuration)
+- [API Reference](#api-reference)
+- [Performance Benchmarks](#performance-benchmarks)
+- [Documentation](#documentation)
+
+---
+
+## Key Features
+
+| Feature | Description |
+|---------|-------------|
+| **4x Faster Training** | Quantum-optimized algorithms with distributed workers |
+| **4x Memory Reduction** | Adaptive 4-16 bit quantization based on weight importance |
+| **Production-Ready** | Byzantine fault tolerance and dynamic load balancing |
+| **Quantum-Inspired** | Superposition aggregation, entanglement simulation, QUBO optimization |
+| **Distributed** | Ray-based MapReduce with O(log n) hierarchical aggregation |
+
+---
+
+## Installation
+
+### From PyPI (Recommended)
+```bash
+pip install hqde
+```
+
+### From Source
+```bash
+git clone https://github.com/Prathmesh333/HQDE-PyPI.git
+cd HQDE-PyPI
+pip install -e .
+```
+
+---
+
+## Quick Start
+
+```python
+from hqde import create_hqde_system
+import torch.nn as nn
+
+# Define your PyTorch model
+class MyModel(nn.Module):
+    def __init__(self, num_classes=10):
+        super().__init__()
+        self.layers = nn.Sequential(
+            nn.Conv2d(3, 32, 3, padding=1),
+            nn.ReLU(),
+            nn.MaxPool2d(2),
+            nn.Conv2d(32, 64, 3, padding=1),
+            nn.ReLU(),
+            nn.AdaptiveAvgPool2d(1),
+            nn.Flatten(),
+            nn.Linear(64, num_classes)
+        )
+
+    def forward(self, x):
+        return self.layers(x)
+
+# Create HQDE system with 4 distributed workers
+hqde_system = create_hqde_system(
+    model_class=MyModel,
+    model_kwargs={'num_classes': 10},
+    num_workers=4
+)
+
+# Train the ensemble
+metrics = hqde_system.train(train_loader, num_epochs=10)
+
+# Make predictions (ensemble voting)
+predictions = hqde_system.predict(test_loader)
+
+# Cleanup resources
+hqde_system.cleanup()
+```
+
+**Examples:**
+```bash
+python examples/quick_start.py           # Quick demo
+python examples/cifar10_synthetic_test.py # CIFAR-10 benchmark
+python examples/cifar10_test.py          # Real CIFAR-10 dataset
+```
+
+---
+
+## Architecture Overview
+
+```
+┌─────────────────────────────────────────────────────────────────┐
+│                    HQDE SYSTEM ARCHITECTURE                      │
+├─────────────────────────────────────────────────────────────────┤
+│                                                                  │
+│  ┌─────────────┐    ┌─────────────────┐    ┌────────────────┐   │
+│  │   QUANTUM   │    │   DISTRIBUTED   │    │    ADAPTIVE    │   │
+│  │  INSPIRED   │───▶│    ENSEMBLE     │───▶│  QUANTIZATION  │   │
+│  │ ALGORITHMS  │    │    LEARNING     │    │                │   │
+│  └─────────────┘    └─────────────────┘    └────────────────┘   │
+│                                                                  │
+└─────────────────────────────────────────────────────────────────┘
+```
+
+### Project Structure
+
+```
+hqde/
+├── core/
+│   └── hqde_system.py           # Main system, workers, quantization
+├── quantum/
+│   ├── quantum_aggregator.py    # Superposition and entanglement
+│   ├── quantum_noise.py         # Quantum noise generation
+│   └── quantum_optimization.py  # QUBO and quantum annealing
+├── distributed/
+│   ├── mapreduce_ensemble.py      # MapReduce pattern
+│   ├── hierarchical_aggregator.py # Tree aggregation
+│   ├── fault_tolerance.py         # Byzantine fault tolerance
+│   └── load_balancer.py           # Dynamic load balancing
+└── utils/
+    └── performance_monitor.py     # System monitoring
+```
+
+---
+
+## Quantum-Inspired Algorithms
+
+**Note:** HQDE uses quantum-inspired algorithms on classical hardware, not actual quantum computers.
+
+### Quantum Superposition Aggregation
+
+Combines ensemble predictions using quantum amplitude-like weights:
+
+```python
+# Confidence scores converted to quantum amplitudes
+amplitudes = sqrt(softmax(confidence_scores))
+
+# Superposition combination
+superposition = sum(amplitude_i * prediction_i)
+```
+
+**Location:** `hqde/quantum/quantum_aggregator.py`
+
+### Entanglement-Based Correlation
+
+Models correlations between ensemble members using an entanglement matrix:
+
+```python
+# Symmetric entanglement matrix
+entanglement_matrix[i,j] = correlation(model_i, model_j) * strength
+
+# Weight models by their entanglement with others
+entangled_weights = softmax(cosine_similarity @ entanglement_matrix)
+```
+
+**Location:** `hqde/quantum/quantum_aggregator.py`
+
+### Quantum Annealing Optimization
+
+Uses QUBO (Quadratic Unconstrained Binary Optimization) for ensemble selection:
+
+```python
+# QUBO formulation for selecting optimal models
+qubo_matrix = formulate_qubo(candidate_models, constraints)
+
+# Solve using simulated quantum annealing
+solution = quantum_annealing_solve(qubo_matrix)
+```
+
+**Location:** `hqde/quantum/quantum_optimization.py`
+
+---
+
+## Distributed Computing
+
+HQDE uses Ray for distributed computing with several patterns:
+
+### Ray Worker Architecture
+
+```python
+# GPUs are automatically divided among workers
+@ray.remote(num_gpus=gpu_per_worker)
+class EnsembleWorker:
+    def train_step(self, data_batch, targets):
+        # Each worker trains its own model copy
+        ...
+```
+
+### MapReduce Weight Aggregation
+
+```
+MAP      →    SHUFFLE    →    REDUCE
+Workers       Group by        Aggregate
+weights       parameter       weights
+              name
+```
+
+**Location:** `hqde/distributed/mapreduce_ensemble.py`
+
+### Hierarchical Tree Aggregation
+
+Communication Complexity: **O(log n)**
+
+```
+Level 0 (Root):           [AGG]
+                         /     \
+Level 1:            [AGG]       [AGG]
+                   /    \       /    \
+Level 2:        [W1]  [W2]   [W3]  [W4]
+```
+
+**Location:** `hqde/distributed/hierarchical_aggregator.py`
+
+### Byzantine Fault Tolerance
+
+Tolerates up to 33% faulty or malicious workers:
+
+- **Outlier Detection:** Median Absolute Deviation (MAD)
+- **Robust Aggregation:** Geometric median
+- **Reliability Tracking:** Source reputation scores
+
+**Location:** `hqde/distributed/fault_tolerance.py`
+
+### Dynamic Load Balancing
+
+Multi-factor node selection scoring:
+- 40% success rate
+- 30% current load
+- 20% execution speed
+- 10% capability match
+
+**Location:** `hqde/distributed/load_balancer.py`
+
+---
+
+## Adaptive Quantization
+
+Dynamically adjusts precision based on weight importance:
+
+| Weight Importance | Bits | Compression |
+|------------------|------|-------------|
+| High (critical)  | 16   | 2x |
+| Medium (default) | 8    | 4x |
+| Low (redundant)  | 4    | 8x |
+
+**Importance Score = 70% × |weight| + 30% × |gradient|**
+
+```python
+quantization_config = {
+    'base_bits': 8,   # Default precision
+    'min_bits': 4,    # High compression for unimportant weights
+    'max_bits': 16    # High precision for critical weights
+}
+```
+
+**Location:** `hqde/core/hqde_system.py`
+
+---
+
+## Configuration
+
+### Full Configuration Example
+
+```python
+from hqde import create_hqde_system
+
+# Quantization settings
+quantization_config = {
+    'base_bits': 8,
+    'min_bits': 4,
+    'max_bits': 16
+}
+
+# Quantum aggregation settings
+aggregation_config = {
+    'noise_scale': 0.005,
+    'exploration_factor': 0.1,
+    'entanglement_strength': 0.1
+}
+
+# Create system
+hqde_system = create_hqde_system(
+    model_class=YourModel,
+    model_kwargs={'num_classes': 10},
+    num_workers=8,
+    quantization_config=quantization_config,
+    aggregation_config=aggregation_config
+)
+```
+
+---
+
+## API Reference
+
+### Core Classes
+
+| Class | Description | Location |
+|-------|-------------|----------|
+| `HQDESystem` | Main entry point | `hqde/core/hqde_system.py` |
+| `DistributedEnsembleManager` | Manages Ray workers | `hqde/core/hqde_system.py` |
+| `AdaptiveQuantizer` | Weight compression | `hqde/core/hqde_system.py` |
+
+### Quantum Classes
+
+| Class | Description | Location |
+|-------|-------------|----------|
+| `QuantumEnsembleAggregator` | Superposition/entanglement aggregation | `hqde/quantum/quantum_aggregator.py` |
+| `QuantumNoiseGenerator` | Exploration noise | `hqde/quantum/quantum_noise.py` |
+| `QuantumEnsembleOptimizer` | QUBO-based selection | `hqde/quantum/quantum_optimization.py` |
+
+### Distributed Classes
+
+| Class | Description | Location |
+|-------|-------------|----------|
+| `MapReduceEnsembleManager` | MapReduce pattern | `hqde/distributed/mapreduce_ensemble.py` |
+| `HierarchicalAggregator` | Tree aggregation | `hqde/distributed/hierarchical_aggregator.py` |
+| `ByzantineFaultTolerantAggregator` | Fault tolerance | `hqde/distributed/fault_tolerance.py` |
+| `DynamicLoadBalancer` | Work distribution | `hqde/distributed/load_balancer.py` |
+
+### Factory Function
+
+```python
+def create_hqde_system(
+    model_class,           # PyTorch model class
+    model_kwargs,          # Model initialization parameters
+    num_workers=4,         # Number of distributed workers
+    quantization_config=None,
+    aggregation_config=None
+) -> HQDESystem
+```
+
+---
+
+## Performance Benchmarks
+
+| Metric | Traditional Ensemble | HQDE | Improvement |
+|--------|---------------------|------|-------------|
+| Memory Usage | 2.4 GB | 0.6 GB | 4x reduction |
+| Training Time | 45 min | 12 min | 3.75x faster |
+| Communication | 800 MB | 100 MB | 8x less data |
+| Test Accuracy | 91.2% | 93.7% | +2.5% |
+
+---
+
+## Documentation
+
+- [HOW_TO_RUN.md](HOW_TO_RUN.md) - Detailed setup and usage guide
+- [docs/](docs/) - Technical documentation
+- [examples/](examples/) - Working code examples
+
+---
+
+## Contributing
+
+1. Fork the repository
+2. Create a feature branch (`git checkout -b feature/new-feature`)
+3. Commit your changes (`git commit -m 'Add new feature'`)
+4. Push to the branch (`git push origin feature/new-feature`)
+5. Open a Pull Request
+
+---
+
+## License
+
+This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
+
+---
+
+## Citation
+
+```bibtex
+@software{hqde2025,
+  title={HQDE: Hierarchical Quantum-Distributed Ensemble Learning},
+  author={Prathamesh Nikam},
+  year={2025},
+  url={https://github.com/Prathmesh333/HQDE-PyPI}
+}
+```
+
+---
+
+## Support
+
+- **Bug Reports:** [Create an issue](https://github.com/Prathmesh333/HQDE-PyPI/issues)
+- **Feature Requests:** [Create an issue](https://github.com/Prathmesh333/HQDE-PyPI/issues)
+- **Questions:** [Start a discussion](https://github.com/Prathmesh333/HQDE-PyPI/issues)

{hqde-0.1.4.dist-info → hqde-0.1.5.dist-info}/RECORD

@@ -2,7 +2,7 @@ hqde/__init__.py,sha256=jxetUxE9gTqHOpxYDx2ZwcJKIkHa7eMIprl9dGuqiBI,1353
 hqde/__main__.py,sha256=6Dozsi53MxYGWL_vFJaH4KuTVJu_RtcD0Tjpn1bGiF0,3054
 hqde/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 hqde/core/__init__.py,sha256=ZLB6uBaJKyfTaSeHckiyW21HUzKcDGo52hdj0gJzL1U,499
-hqde/core/hqde_system.py,sha256=jrT4tlz8eusSf-2EkTCEvloXvZYjOUfMl8NZBmBoGYc,19851
+hqde/core/hqde_system.py,sha256=D2-26bvuwQHKynIRAR9-yAY4hER-WcEwenUbblasf4A,22273
 hqde/distributed/__init__.py,sha256=qOzxRxTJejXGiNwv2Ibts5m4pSLt8KtzLWu0RgEQnuU,584
 hqde/distributed/fault_tolerance.py,sha256=TMfLCXL14BO0TYL834r41oKoZ9dxxTp99Ux1d6hBMfw,14801
 hqde/distributed/hierarchical_aggregator.py,sha256=UbtB2qU1ws70594woK_bJhvbjN6PA9XAWxggT8F00rY,15790
@@ -17,8 +17,8 @@ hqde/utils/config_manager.py,sha256=GY_uFBwj6qJ_ESkopIjR_vQwLIcILNqdNj2o_GFFAdg,
 hqde/utils/data_utils.py,sha256=2CVHULh45Usf9zcvM7i3qeZkpLNzRSEPDQ4vCjHk14E,264
 hqde/utils/performance_monitor.py,sha256=J4VntvwnBwMRAArtuVDr13oKcVjr4y5WWowW1dm21rI,16644
 hqde/utils/visualization.py,sha256=NwiUrgMQFBeqrIblp2qFWl71bFNG58FZKESK2-GB8eM,185
-hqde-0.1.4.dist-info/licenses/LICENSE,sha256=ACTIUEzMwldWiL-H94KKJaGyUNxu_L5EQylXnagPamE,1065
-hqde-0.1.4.dist-info/METADATA,sha256=aQjNKhBr3StdtoB_rr5J4RSIpNtdiOeiOD1GstI6chg,7887
-hqde-0.1.4.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-hqde-0.1.4.dist-info/top_level.txt,sha256=lDNw5jGWRhvYQohaYu7Cm4F7vd3YFPIwoLULxJNopqc,5
-hqde-0.1.4.dist-info/RECORD,,
+hqde-0.1.5.dist-info/licenses/LICENSE,sha256=ACTIUEzMwldWiL-H94KKJaGyUNxu_L5EQylXnagPamE,1065
+hqde-0.1.5.dist-info/METADATA,sha256=w5yNPRW_SEM6Qn3UhKWdaxv8b6TEPkwHK-MKGDi97_Y,13989
+hqde-0.1.5.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+hqde-0.1.5.dist-info/top_level.txt,sha256=lDNw5jGWRhvYQohaYu7Cm4F7vd3YFPIwoLULxJNopqc,5
+hqde-0.1.5.dist-info/RECORD,,

{hqde-0.1.4.dist-info → hqde-0.1.5.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (80.9.0)
+Generator: setuptools (80.10.2)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

hqde-0.1.4.dist-info/METADATA

@@ -1,237 +0,0 @@
-Metadata-Version: 2.4
-Name: hqde
-Version: 0.1.4
-Summary: Hierarchical Quantum-Distributed Ensemble Learning Framework
-Author-email: HQDE Team <hqde@example.com>
-Maintainer-email: HQDE Team <hqde@example.com>
-License: MIT
-Project-URL: Homepage, https://github.com/Prathmesh333/Hierarchical-Quantum-Distributed-Ensemble-Learning
-Project-URL: Repository, https://github.com/Prathmesh333/Hierarchical-Quantum-Distributed-Ensemble-Learning
-Project-URL: Documentation, https://github.com/Prathmesh333/Hierarchical-Quantum-Distributed-Ensemble-Learning/blob/main/HOW_TO_RUN.md
-Project-URL: Bug Reports, https://github.com/Prathmesh333/Hierarchical-Quantum-Distributed-Ensemble-Learning/issues
-Keywords: machine-learning,quantum,distributed,ensemble,deep-learning,pytorch,ray
-Classifier: Development Status :: 4 - Beta
-Classifier: Intended Audience :: Science/Research
-Classifier: Intended Audience :: Developers
-Classifier: License :: OSI Approved :: MIT License
-Classifier: Operating System :: OS Independent
-Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.9
-Classifier: Programming Language :: Python :: 3.10
-Classifier: Programming Language :: Python :: 3.11
-Classifier: Programming Language :: Python :: 3.12
-Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Classifier: Topic :: Software Development :: Libraries :: Python Modules
-Requires-Python: >=3.9
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: numpy>=2.0.2
-Requires-Dist: torch>=2.8.0
-Requires-Dist: torchvision>=0.23.0
-Requires-Dist: ray[default]>=2.49.2
-Requires-Dist: scikit-learn>=1.6.1
-Requires-Dist: psutil>=7.1.0
-Provides-Extra: dev
-Requires-Dist: pytest>=8.4.2; extra == "dev"
-Requires-Dist: ipython>=8.18.1; extra == "dev"
-Requires-Dist: matplotlib>=3.9.4; extra == "dev"
-Requires-Dist: pandas>=2.3.2; extra == "dev"
-Requires-Dist: seaborn>=0.13.2; extra == "dev"
-Requires-Dist: tqdm>=4.67.1; extra == "dev"
-Dynamic: license-file
-
-# HQDE - Hierarchical Quantum-Distributed Ensemble Learning
-
-[![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/)
-[![PyTorch](https://img.shields.io/badge/PyTorch-2.8+-red.svg)](https://pytorch.org/)
-[![Ray](https://img.shields.io/badge/Ray-2.49+-green.svg)](https://ray.io/)
-[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
-
-> **A production-ready framework for distributed ensemble learning with quantum-inspired algorithms and adaptive quantization.**
-
-HQDE combines cutting-edge quantum-inspired algorithms with distributed computing to deliver superior machine learning performance with significantly reduced memory usage and training time.
-
-## ✨ Why HQDE?
-
-- **🚀 4x faster training** with quantum-optimized algorithms
-- **💾 4x memory reduction** through adaptive quantization
-- **🔧 Production-ready** with fault tolerance and load balancing
-- **🧠 Quantum-inspired** ensemble aggregation methods
-- **🌐 Distributed** processing with automatic scaling
-
-## 📦 Installation
-
-### Option 1: Install from PyPI (Recommended)
-```bash
-pip install hqde
-```
-
-### Option 2: Install from Source
-```bash
-git clone https://github.com/Prathmesh333/HQDE-PyPI.git
-cd HQDE-PyPI
-pip install -e .
-```
-
-## 🚀 Quick Start
-
-```python
-from hqde import create_hqde_system
-import torch.nn as nn
-
-# Define your PyTorch model
-class MyModel(nn.Module):
-    def __init__(self, num_classes=10):
-        super().__init__()
-        self.layers = nn.Sequential(
-            nn.Conv2d(3, 32, 3, padding=1),
-            nn.ReLU(),
-            nn.MaxPool2d(2),
-            nn.Conv2d(32, 64, 3, padding=1),
-            nn.ReLU(),
-            nn.AdaptiveAvgPool2d(1),
-            nn.Flatten(),
-            nn.Linear(64, num_classes)
-        )
-
-    def forward(self, x):
-        return self.layers(x)
-
-# Create HQDE system (it's that simple!)
-hqde_system = create_hqde_system(
-    model_class=MyModel,
-    model_kwargs={'num_classes': 10},
-    num_workers=4  # Use 4 distributed workers
-)
-
-# Train your ensemble
-metrics = hqde_system.train(train_loader, num_epochs=10)
-
-# Make predictions
-predictions = hqde_system.predict(test_loader)
-```
-
-## 🧪 Try the Examples
-
-```bash
-# Quick demo (30 seconds)
-python examples/quick_start.py
-
-# CIFAR-10 benchmark test
-python examples/cifar10_synthetic_test.py
-
-# Real CIFAR-10 dataset
-python examples/cifar10_test.py
-```
-
-### Expected Results
-```
-=== HQDE CIFAR-10 Test Results ===
-Training Time: 18.29 seconds
-Test Accuracy: 86.10%
-Memory Usage: 0.094 MB
-Ensemble Diversity: 96.8%
-```
-
-## ⚙️ Key Features
-
-### 🧠 Quantum-Inspired Algorithms
-- **Quantum Superposition Aggregation**: Advanced ensemble combination
-- **Entanglement-Based Correlation**: Sophisticated member coordination
-- **Quantum Noise Injection**: Enhanced exploration and generalization
-
-### 📊 Adaptive Quantization
-- **Dynamic Bit Allocation**: 4-16 bit precision based on importance
-- **Real-time Optimization**: Automatic compression without accuracy loss
-- **Memory Efficiency**: Up to 20x reduction vs traditional methods
-
-### 🌐 Distributed Processing
-- **MapReduce Architecture**: Scalable with Ray framework
-- **Byzantine Fault Tolerance**: Robust against node failures
-- **Hierarchical Aggregation**: O(log n) communication complexity
-
-## 📈 Performance Benchmarks
-
-| Metric | Traditional Ensemble | HQDE | Improvement |
-|--------|---------------------|------|-------------|
-| Memory Usage | 2.4 GB | 0.6 GB | **4x reduction** |
-| Training Time | 45 min | 12 min | **3.75x faster** |
-| Communication | 800 MB | 100 MB | **8x less data** |
-| Test Accuracy | 91.2% | 93.7% | **+2.5% better** |
-
-## 🔧 Configuration
-
-Customize HQDE for your needs:
-
-```python
-# Fine-tune quantization
-quantization_config = {
-    'base_bits': 8,      # Default precision
-    'min_bits': 4,       # High compression
-    'max_bits': 16       # High precision
-}
-
-# Adjust quantum parameters
-aggregation_config = {
-    'noise_scale': 0.005,           # Quantum noise level
-    'exploration_factor': 0.1,      # Exploration strength
-    'entanglement_strength': 0.1    # Ensemble correlation
-}
-
-# Scale distributed processing
-hqde_system = create_hqde_system(
-    model_class=YourModel,
-    num_workers=8,  # Scale up for larger datasets
-    quantization_config=quantization_config,
-    aggregation_config=aggregation_config
-)
-```
-
-## 📚 Documentation
-
-- **[HOW_TO_RUN.md](HOW_TO_RUN.md)** - Detailed setup and usage guide
-- **[Examples](examples/)** - Working code examples and demos
-- **[API Reference](hqde/)** - Complete module documentation
-
-## 🤝 Contributing
-
-We welcome contributions! Please:
-
-1. Fork the repository
-2. Create a feature branch (`git checkout -b feature/amazing-feature`)
-3. Commit your changes (`git commit -m 'Add amazing feature'`)
-4. Push to the branch (`git push origin feature/amazing-feature`)
-5. Open a Pull Request
-
-## 📄 License
-
-This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
-
-## 🔗 Citation
-
-If you use HQDE in your research, please cite:
-
-```bibtex
-@software{hqde2025,
-  title={HQDE: Hierarchical Quantum-Distributed Ensemble Learning},
-  author={Prathamesh Nikam},
-  year={2025},
-  url={https://github.com/Prathmesh333/HQDE-PyPI}
-}
-```
-
-## 🆘 Support
-
-- **🐛 Bug Reports**: [Create an issue](https://github.com/Prathmesh333/HQDE-PyPI/issues)
-- **💡 Feature Requests**: [Create an issue](https://github.com/Prathmesh333/HQDE-PyPI/issues)
-- **💬 Questions**: [Start a discussion](https://github.com/Prathmesh333/HQDE-PyPI/issues)
-
----
-
-<div align="center">
-
-**Built with ❤️ for the machine learning community**
-
-[⭐ Star](https://github.com/Prathmesh333/HQDE-PyPI/stargazers) • [🍴 Fork](https://github.com/Prathmesh333/HQDE-PyPI/fork) • [📝 Issues](https://github.com/Prathmesh333/HQDE-PyPI/issues)
-
-</div>