Delete dataset/legacy_enhanced_data

The telemetry code needs to be rework on

dataset/legacy_enhanced_data/compare_legacy_vs_v2.py

@@ -1,68 +0,0 @@
-
-# Compare YOUR legacy data with v2 telemetry data
-import matplotlib.pyplot as plt
-import seaborn as sns
-
-def compare_legacy_vs_v2():
-    """Compare legacy trading data with v2 telemetry"""
-    
-    # Load legacy data
-    legacy_df = load_legacy_data()
-    
-    # Load v2 data (current dataset)
-    from datasets import load_dataset
-    try:
-        v2_ds = load_dataset("rmems/Spikenaut-SNN-v2-Telemetry-Data-Weights-Parameters")
-        v2_df = v2_ds['train'].to_pandas()
-        print("✅ V2 dataset loaded")
-    except:
-        print("⚠️ V2 dataset not available, using sample")
-        v2_df = None
-    
-    print("\n🔍 Dataset Comparison:")
-    print(f"Legacy: {len(legacy_df):,} records (trading focus)")
-    if v2_df is not None:
-        print(f"V2: {len(v2_df)} records (telemetry focus)")
-    
-    # Compare time ranges
-    if 'timestamp' in legacy_df.columns:
-        legacy_df['timestamp'] = pd.to_datetime(legacy_df['timestamp'])
-        print(f"\n⏰ Time Coverage:")
-        print(f"Legacy: {legacy_df['timestamp'].min()} to {legacy_df['timestamp'].max()}")
-        print(f"Duration: {legacy_df['timestamp'].max() - legacy_df['timestamp'].min()}")
-    
-    # Compare data types
-    print(f"\n📋 Data Types:")
-    print(f"Legacy focus: Trading actions, portfolio management, blockchain metrics")
-    if v2_df is not None:
-        print(f"V2 focus: Blockchain telemetry, spike encodings, SNN features")
-    
-    # Visualize portfolio evolution (legacy)
-    if 'portfolio_value' in legacy_df.columns:
-        plt.figure(figsize=(12, 4))
-        
-        plt.subplot(1, 2, 1)
-        # Sample every 1000th point for performance
-        sample_legacy = legacy_df.iloc[::1000]
-        plt.plot(sample_legacy.index, sample_legacy['portfolio_value'], alpha=0.7)
-        plt.title('🦁 Legacy Portfolio Evolution')
-        plt.xlabel('Record Index')
-        plt.ylabel('Portfolio Value ($)')
-        plt.grid(True, alpha=0.3)
-        
-        # Action distribution
-        plt.subplot(1, 2, 2)
-        action_counts = legacy_df['action'].value_counts()
-        plt.pie(action_counts.values, labels=action_counts.index, autopct='%1.1f%%')
-        plt.title('Legacy Action Distribution')
-        
-        plt.tight_layout()
-        plt.show()
-    
-    print("\n🎯 Key Insights:")
-    print("• Legacy: Rich trading history with 200K+ records")
-    print("• V2: Focused telemetry with spike encodings")
-    print("• Combined: Complete picture of Spikenaut evolution")
-
-# Run comparison
-compare_legacy_vs_v2()

dataset/legacy_enhanced_data/legacy_summary_statistics.json

@@ -1,41 +0,0 @@
-{
-  "legacy_dataset_info": {
-    "total_records": 223020,
-    "file_size_mb": 182.3,
-    "date_range": {
-      "start": "2026-03-12T06:31:49.460483249+00:00",
-      "end": "2026-03-15T14:08:16.650911711+00:00"
-    },
-    "processing_date": "2026-03-23T07:13:53.008746"
-  },
-  "data_quality": {
-    "valid_json_rate": 100.0,
-    "completeness": {
-      "timestamp": 100.0,
-      "action": 100.0,
-      "portfolio_value": 100.0,
-      "price_usd": 100.0
-    }
-  },
-  "trading_metrics": {
-    "total_actions": 10000,
-    "observe_actions": 9936,
-    "buy_actions": 29,
-    "sell_actions": 35,
-    "portfolio_value_range": {
-      "min": 500.0,
-      "max": 1102.5507,
-      "mean": 990.2608183219999
-    }
-  },
-  "blockchain_metrics": {
-    "quai_block_utilization": {
-      "mean": 0.65,
-      "std": 0.0
-    },
-    "quai_gas_price": {
-      "mean": 10.0,
-      "std": 0.0
-    }
-  }
-}

dataset/legacy_enhanced_data/load_legacy_data.py

@@ -1,59 +0,0 @@
-
-# Load and analyze YOUR massive legacy Spikenaut dataset
-import json
-import pandas as pd
-import numpy as np
-from pathlib import Path
-
-def load_legacy_data(chunk_dir="legacy_enhanced_data"):
-    """Load your enhanced legacy dataset"""
-    all_data = []
-    
-    chunk_dir = Path(chunk_dir)
-    chunk_files = sorted(chunk_dir.glob("legacy_chunk_*.jsonl"))
-    
-    print(f"🦁 Loading {len(chunk_files)} legacy data chunks...")
-    
-    for chunk_file in chunk_files:
-        with open(chunk_file, 'r') as f:
-            for line in f:
-                if line.strip():
-                    record = json.loads(line)
-                    all_data.append(record)
-    
-    df = pd.DataFrame(all_data)
-    print(f"✅ Loaded {len(df):,} records from legacy dataset")
-    
-    return df
-
-# Load your legacy data
-legacy_df = load_legacy_data()
-
-print("\n📊 Legacy Dataset Overview:")
-print(f"  Records: {len(legacy_df):,}")
-print(f"  Columns: {list(legacy_df.columns)}")
-print(f"  Date range: {legacy_df['timestamp'].min()} to {legacy_df['timestamp'].max()}")
-
-# Analyze trading patterns
-print("\n💰 Trading Analysis:")
-action_counts = legacy_df['action'].value_counts()
-for action, count in action_counts.items():
-    print(f"  {action}: {count:,} ({count/len(legacy_df)*100:.1f}%)")
-
-# Portfolio performance over time
-if 'portfolio_value' in legacy_df.columns:
-    portfolio_stats = legacy_df['portfolio_value'].describe()
-    print(f"\n📈 Portfolio Performance:")
-    print(f"  Initial: ${portfolio_stats['min']:.2f}")
-    print(f"  Final: ${portfolio_stats['max']:.2f}")
-    print(f"  Mean: ${portfolio_stats['mean']:.2f}")
-    print(f"  Return: {(portfolio_stats['max']/500 - 1)*100:.2f}%")
-
-# Blockchain health analysis
-if 'blockchain_health_score' in legacy_df.columns:
-    health_stats = legacy_df['blockchain_health_score'].describe()
-    print(f"\n⛓️ Blockchain Health:")
-    print(f"  Mean score: {health_stats['mean']:.3f}")
-    print(f"  Health trend: {'Improving' if health_stats['mean'] > 0.6 else 'Stable' if health_stats['mean'] > 0.4 else 'Declining'}")
-
-print("\n🎉 Your legacy dataset shows rich trading and blockchain telemetry!")