Spaces:
Sleeping
Sleeping
File size: 8,566 Bytes
c8df794 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 | """
Memory usage monitoring and optimization script
Helps monitor and optimize RAM usage for the disease detection system
"""
import psutil
import torch
import gc
import os
import time
from pathlib import Path
class MemoryMonitor:
"""Monitor and optimize memory usage"""
def __init__(self):
self.process = psutil.Process(os.getpid())
self.initial_memory = self.get_memory_usage()
self.peak_memory = self.initial_memory
def get_memory_usage(self):
"""Get current memory usage in MB"""
return self.process.memory_info().rss / 1024 / 1024
def get_memory_percent(self):
"""Get memory usage as percentage of total system memory"""
return self.process.memory_percent()
def get_system_memory_info(self):
"""Get system memory information"""
memory = psutil.virtual_memory()
return {
"total_gb": memory.total / 1024 / 1024 / 1024,
"available_gb": memory.available / 1024 / 1024 / 1024,
"used_percent": memory.percent,
"free_gb": memory.free / 1024 / 1024 / 1024
}
def optimize_memory(self):
"""Force memory optimization"""
gc.collect()
if torch.cuda.is_available():
torch.cuda.empty_cache()
def check_memory_status(self):
"""Check if memory usage is within acceptable limits"""
current = self.get_memory_usage()
self.peak_memory = max(self.peak_memory, current)
status = {
"current_mb": current,
"peak_mb": self.peak_memory,
"increase_mb": current - self.initial_memory,
"within_512mb": current < 512,
"system_percent": self.get_memory_percent(),
"status": "OK" if current < 512 else "HIGH"
}
return status
def print_memory_report(self):
"""Print detailed memory report"""
status = self.check_memory_status()
system_info = self.get_system_memory_info()
print("\n" + "="*50)
print("MEMORY USAGE REPORT")
print("="*50)
print(f"Process Memory Usage:")
print(f" Current: {status['current_mb']:.1f} MB")
print(f" Peak: {status['peak_mb']:.1f} MB")
print(f" Increase: {status['increase_mb']:.1f} MB")
print(f" Status: {status['status']}")
print(f" Within 512MB limit: {'✅' if status['within_512mb'] else '❌'}")
print(f"\nSystem Memory:")
print(f" Total: {system_info['total_gb']:.1f} GB")
print(f" Available: {system_info['available_gb']:.1f} GB")
print(f" Used: {system_info['used_percent']:.1f}%")
if torch.cuda.is_available():
gpu_memory = torch.cuda.memory_allocated() / 1024 / 1024
gpu_cached = torch.cuda.memory_reserved() / 1024 / 1024
print(f"\nGPU Memory:")
print(f" Allocated: {gpu_memory:.1f} MB")
print(f" Cached: {gpu_cached:.1f} MB")
print("="*50)
return status
def test_model_memory_usage():
"""Test memory usage of different model configurations"""
monitor = MemoryMonitor()
print("Testing memory usage of disease detection models...")
monitor.print_memory_report()
try:
# Test importing modules
print("\n1. Testing module imports...")
import sys
sys.path.append('src')
from src.model_lite import CropDiseaseResNet50Lite, TinyDiseaseClassifier
monitor.optimize_memory()
status = monitor.check_memory_status()
print(f"After imports: {status['current_mb']:.1f} MB")
# Test lite model
print("\n2. Testing CropDiseaseResNet50Lite...")
model_lite = CropDiseaseResNet50Lite(15, pretrained=False)
status = monitor.check_memory_status()
print(f"After lite model creation: {status['current_mb']:.1f} MB")
print(f"Model size: {model_lite.get_model_size():.1f} MB")
# Test with dummy input
dummy_input = torch.randn(1, 3, 224, 224)
with torch.no_grad():
output = model_lite(dummy_input)
status = monitor.check_memory_status()
print(f"After inference: {status['current_mb']:.1f} MB")
del model_lite, dummy_input, output
monitor.optimize_memory()
# Test tiny model
print("\n3. Testing TinyDiseaseClassifier...")
model_tiny = TinyDiseaseClassifier(15)
status = monitor.check_memory_status()
print(f"After tiny model creation: {status['current_mb']:.1f} MB")
print(f"Model size: {model_tiny.get_model_size():.1f} MB")
# Test with dummy input
dummy_input = torch.randn(1, 3, 224, 224)
with torch.no_grad():
output = model_tiny(dummy_input)
status = monitor.check_memory_status()
print(f"After inference: {status['current_mb']:.1f} MB")
del model_tiny, dummy_input, output
monitor.optimize_memory()
except Exception as e:
print(f"Error during testing: {e}")
print("\n4. Final memory report:")
monitor.print_memory_report()
return monitor
def benchmark_memory_usage():
"""Benchmark memory usage with continuous monitoring"""
monitor = MemoryMonitor()
print("Starting memory benchmark...")
try:
# Simulate API startup
print("Simulating API startup...")
import sys
sys.path.append('src')
from src.model_lite import CropDiseaseResNet50Lite
from src.explain_lite import CropDiseaseExplainerLite
# Load model
model = CropDiseaseResNet50Lite(15, pretrained=False)
model.eval()
# Initialize explainer
explainer = CropDiseaseExplainerLite(model, ["class1", "class2"], 'cpu')
status = monitor.check_memory_status()
print(f"After model loading: {status['current_mb']:.1f} MB")
# Simulate multiple predictions
print("Simulating predictions...")
for i in range(5):
dummy_input = torch.randn(1, 3, 224, 224)
with torch.no_grad():
output = model(dummy_input)
# Cleanup after each prediction
del dummy_input, output
monitor.optimize_memory()
status = monitor.check_memory_status()
print(f"Prediction {i+1}: {status['current_mb']:.1f} MB")
print("\nFinal benchmark results:")
monitor.print_memory_report()
except Exception as e:
print(f"Benchmark error: {e}")
monitor.print_memory_report()
def get_optimization_recommendations():
"""Get memory optimization recommendations"""
monitor = MemoryMonitor()
system_info = monitor.get_system_memory_info()
recommendations = []
if system_info["total_gb"] < 2:
recommendations.append("🔴 Very low system memory. Use TinyDiseaseClassifier instead of ResNet50")
recommendations.append("🔴 Disable explanations completely")
recommendations.append("🔴 Set max_workers=1 for API")
elif system_info["total_gb"] < 4:
recommendations.append("🟡 Low system memory. Use CropDiseaseResNet50Lite")
recommendations.append("🟡 Enable memory efficient mode")
recommendations.append("🟡 Limit concurrent requests to 2")
else:
recommendations.append("🟢 Sufficient system memory for standard operation")
recommendations.append("🟢 Can use full model with optimizations")
if system_info["used_percent"] > 80:
recommendations.append("⚠️ High system memory usage. Close other applications")
print("\nMEMORY OPTIMIZATION RECOMMENDATIONS:")
print("="*50)
for rec in recommendations:
print(rec)
print("="*50)
return recommendations
if __name__ == "__main__":
print("Disease Detection Memory Monitor")
print("="*50)
# Run tests
print("1. Testing model memory usage...")
test_model_memory_usage()
print("\n2. Running memory benchmark...")
benchmark_memory_usage()
print("\n3. Getting optimization recommendations...")
get_optimization_recommendations()
print("\nMemory monitoring complete!") |