PyTorch torch.cuda.memory_allocated function
Pytorch torch Reference Manual
torch.cuda.memory_allocatedis a function in PyTorch used to get the allocated GPU memory. It returns the size of memory allocated on the current CUDA device (in bytes).
Function Definition
torch.cuda.memory_allocated(device=None)
Usage Example
Example
import torch
# Check allocated GPU memory
if torch.cuda.is_available():
print(f"Initial memory: {torch.cuda.memory_allocated()} bytes")
# Create tensor
x = torch.randn(1000, 1000).cuda()
print(f"After allocation: {torch.cuda.memory_allocated()} bytes")
# Delete tensor
del x
print(f"After deletion: {torch.cuda.memory_allocated()} bytes")
else:
print("CUDA not available")
# Check allocated GPU memory
if torch.cuda.is_available():
print(f"Initial memory: {torch.cuda.memory_allocated()} bytes")
# Create tensor
x = torch.randn(1000, 1000).cuda()
print(f"After allocation: {torch.cuda.memory_allocated()} bytes")
# Delete tensor
del x
print(f"After deletion: {torch.cuda.memory_allocated()} bytes")
else:
print("CUDA not available")
Other Extensions