bigram.py

import torch
import torch.nn
from torch.nn import functional as F

batch_size = 32
block_size = 8
max_iters = 10_000
eval_interval = 340
learning_rate = 1e-2
device = 'mps' if torch.backends.mps.is_available() else 'cpu'
eval_iters = 200


torch.manual_seed(1337)

# wget https://raw.githubusercontent.com/karpathy/char-rnn/master/data/tinyshakespeare/input.txt

with open("input.txt", "r", encoding="utf-8") as f:
    text = f.read()

chars = sorted(set(text))
vocab_size = len(chars)

s_to_i = {ch: i for i, ch in enumerate(chars)}
i_to_s = {i: ch for i, ch in enumerate(chars)}
encode = lambda s: list(map(s_to_i.get, s))
decode = lambda l: "".join(map(i_to_s.get, l))

data = torch.tensor(encode(text), dtype=torch.long)
n = int(0.9 * len(data))
train_data = data[:n]
val_data = data[n:]

def get_batch(split):
    data = train_data if split == "train" else val_data
    ix = torch.randint(len(data) - block_size, (batch_size,))
    x = torch.stack([data[i: i + block_size] for i in ix])
    y = torch.stack([data[i + 1: i + block_size + 1] for i in ix])
    return x.to(device), y.to(device)

@torch.no_grad()
def estimate_loss():
    out = {}
    model.eval()
    for split in ["train", "val"]:
        losses = torch.zeros(eval_iters)
        for k in range(eval_iters):
            X, Y = get_batch(split)
            logits, loss = model(X, Y)
            losses[k] = loss.item()
        out[split] = losses.mean()
    model.train()
    return out

class BigramModel(torch.nn.Module):
    def __init__(self, vocab_size):
        super().__init__()
        self.token_embedding_table = torch.nn.Embedding(vocab_size, vocab_size)

    def forward(self, idx, targets=None):
        logits = self.token_embedding_table(idx)

        if targets is None:
            loss = None
        else:
            B, T, C = logits.shape
            logits = logits.view(B * T, C)
            targets = targets.view(B * T)
            loss = F.cross_entropy(logits, targets)
        return logits, loss
    
    def generate(self, idx, max_new_tokens):
        for _ in range(max_new_tokens):
            logits, _ = self(idx)
            logits = logits[:, -1, :]
            probs = F.softmax(logits, dim=-1)
            idx_next = torch.multinomial(probs, num_samples=1)
            idx = torch.cat((idx, idx_next), dim=1)
        return idx
    
model = BigramModel(vocab_size)
m = model.to(device)

optimizer = torch.optim.AdamW(model.parameters(), lr=learning_rate)

for i in range(max_iters):
    if i % eval_interval == 0:
        losses = estimate_loss()
        print(f'step {i}: train loss {losses["train"]:.4f}, val loss {losses["val"]:.4f}')

    xb, yb = get_batch("train")
    logits, loss = model(xb, yb)
    optimizer.zero_grad(set_to_none=True)
    loss.backward()
    optimizer.step()

context = torch.zeros((1, 1), dtype=torch.long, device=device)
print(decode(m.generate(context, max_new_tokens=500)[0].tolist()))