Learning Center

import torch
import platform
import time
import sys

def verify_gaashub():
    print("\n" + "="*50)
    print("🚀 GAASHUB SETUP VERIFICATION")
    print("="*50)
    
    # 1. Remote Environment Check
    print(f"\n📍 REMOTE NODE: {platform.node()}")
    print(f"🐧 SYSTEM:     {platform.system()} {platform.release()}")
    
    # 2. GPU Detection
    if not torch.cuda.is_available():
        print("\n❌ CRITICAL: No GPU detected!")
        print("   This script is running on CPU. Check your GaasHub connection.")
        sys.exit(1)

    gpu_name = torch.cuda.get_device_name(0)
    vram_gb = torch.cuda.get_device_properties(0).total_memory / 1e9
    print(f"\n✅ GPU DETECTED: {gpu_name}")
    print(f"   VRAM Available: {vram_gb:.2f} GB")
    
    # 3. Compute Stress Test
    print("\n⚡ Running Compute Stress Test (Matrix Mult)...")
    try:
        start_time = time.time()
        # Generate large matrices on GPU
        a = torch.randn(8000, 8000, device='cuda')
        b = torch.randn(8000, 8000, device='cuda')
        # Perform heavy calculation
        c = torch.matmul(a, b)
        torch.cuda.synchronize() # Ensure operation finishes
        duration = time.time() - start_time
        
        print(f"   Status: SUCCESS")
        print(f"   Execution Time: {duration:.4f} seconds")
        
    except Exception as e:
        print(f"❌ COMPUTE FAILED: {str(e)}")
        sys.exit(1)

    print("\n" + "="*50)
    print("✨ YOUR SETUP IS CORRECT! READY FOR ML TASKS.")
    print("="*50 + "\n")

if __name__ == "__main__":
    verify_gaashub()