Built to make you extraordinarily productive, Cursor is the best way to code with AI.

import torch
import torch.nn as nn
- from torch.utils.data import DataLoader
+ from torch.utils.data import DataLoader, random_split
- from torchvision import datasets
+ from torchvision import datasets, transforms
+ from tqdm import tqdm
+ import yaml
+ from pathlib import Path
+ import json

- def get_dataloaders(batch_size=64):
-   transform = transforms.Compose([transforms.ToTensor()])
-   train = datasets.MNIST(root="data", train=True, download=True, transform=transform)
-   test = datasets.MNIST(root="data", train=False, download=True, transform=transform)
-   return DataLoader(train, batch_size=batch_size, shuffle=True), DataLoader(test, batch_size=batch_size)
+ def load_config(config_path="experiments/config.yaml"):
+   with open(config_path) as f:
+     return yaml.safe_load(f)

class MLP(nn.Module):
-   def __init__(self, hidden=128):
+   def __init__(self, config):
    super().__init__()
+     hidden = config['model']['hidden_size']
+     dropout = config['model']['dropout']

    self.net = nn.Sequential(
-       nn.Linear(hidden, 10),
+       nn.Linear(hidden // 2, 10),
    )

+       # Seed for reproducibility
+       torch.manual_seed(42)
+       if device == "cuda":
+         torch.cuda.manual_seed_all(42)
+       # AMP + Scheduler
+       scaler = torch.cuda.amp.GradScaler(enabled=(device=="cuda"))
+       scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(opt, T_max=epochs)