How do you move tensors and models between CPU and GPU in PyTorch?

PyTorch's device abstraction allows the same code to run on CPU or GPU with minimal changes. The fundamental rule: a model and its input tensors must reside on the same device before any computation, or PyTorch raises a RuntimeError.

import torch
import torch.nn as nn

# Device-agnostic pattern — always write code this way
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print(f"Using device: {device}")

# Move a model to the device
model = nn.Linear(10, 1).to(device)

# Move data to the same device, every batch, inside the loop
for X_batch, y_batch in loader:
    X_batch = X_batch.to(device, non_blocking=True)
    y_batch = y_batch.to(device, non_blocking=True)
    pred = model(X_batch)   # works — both on same device

# WRONG — mismatched devices raises RuntimeError
# model_cpu = nn.Linear(10, 1)              # stays on CPU
# x_gpu = torch.randn(4, 10).to("cuda")
# model_cpu(x_gpu)  # RuntimeError: Expected all tensors on same device

# Checking tensor device
t = torch.randn(3)
print(t.device)            # cpu
t_gpu = t.cuda()           # or t.to("cuda:0")
print(t_gpu.device)        # cuda:0

# GPU memory diagnostics
if torch.cuda.is_available():
    print(torch.cuda.memory_allocated() / 1e9, "GB allocated")
    print(torch.cuda.max_memory_allocated() / 1e9, "GB peak")
    torch.cuda.empty_cache()   # release unused cached memory

# Moving a tensor back to CPU (required before .numpy())
result = t_gpu.cpu().numpy()   # numpy() requires a CPU tensor

# Apple Silicon (M1/M2/M3) GPU support
mps_device = torch.device("mps" if torch.backends.mps.is_available() else "cpu")