gpt.py

import time
import humanize
import torch 
import torch.nn as nn
from torch.nn import functional as F 
torch.manual_seed(1337)

# hyperparameters
batch_size = 16 # 64 
block_size = 8 # 256
max_iters = 0 # 10000
eval_interval = 100 
learning_rate = 3e-4 
device = 'cpu'
eval_iters = 200
n_embd = 16 # 384
n_head = 4 # 6
n_layer = 1 # 6
dropout = 0.2
assert block_size % n_head == 0, "block_size must be divisible by n_head"
# ---------------- 

# wget https://raw.githubusercontent.com/karpathy/char-rnn/master/data/tinyshakespeare/input.txt
with open('input.txt', 'r', encoding='utf8') as f:
    text = f.read()

# vocabulary
chars = sorted(list(set(text)))
vocab_size = len(chars)

# create mapping from characters to integers and vice versa
stoi = {ch:i for i,ch in enumerate(chars)}
itos = {i:ch for i,ch in enumerate(chars)}
encode = lambda x: [stoi[ch] for ch in x]
decode = lambda x: ''.join([itos[i] for i in x])

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

# data loading 
def get_batch(split):
    # generate a small batch of data of inputs x and targets y 
    data = train_data if split == 'train' else val_data
    # randomly sample a bunch of block_size length sequences
    ix = torch.randint(len(data) - block_size, (batch_size,))
    # the sequence
    x = torch.stack([data[i:i+block_size] for i in ix]) 
    # the target (next character)
    y = torch.stack([data[i+1:i+block_size+1] for i in ix])
    x, y = x.to(device), y.to(device)
    return x, y

@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 Head(nn.Module):
    """one head of self-attention"""

    def __init__(self, head_size):
        super().__init__()
        self.key = nn.Linear(n_embd, head_size, bias=False)
        self.query = nn.Linear(n_embd, head_size, bias=False)
        self.value = nn.Linear(n_embd, head_size, bias=False)
        self.register_buffer('tril', torch.tril(torch.ones(block_size, block_size)))
        self.dropout = nn.Dropout(dropout)

    def forward(self, x):
        B,T,C = x.shape 
        k = self.key(x) # (B,T,C)
        q = self.query(x) # (B,T,C)

        # compute attention scores ("affinities")
        wei = q @ k.transpose(-2,-1) * C**-0.5 # (B,T,C) @ (B,C,T) = (B,T,T)
        wei = wei.masked_fill(self.tril[:T, :T] ==0, float('-inf')) # (B, T, T)
        wei = F.softmax(wei, dim=-1) # (B, T, T)
        wei = self.dropout(wei)

        # perform the weighted aggregation of the values 
        v = self.value(x) # (B, T, C)
        out = wei @ v # (B, T, T) @ (B, T, C) -> (B, T, C)
        return out

class MultiHeadAttention(nn.Module):
    """multiple heads of self-attention in paralell"""
    def __init__(self, num_heads, head_size):
        super().__init__()
        self.heads = nn.ModuleList([Head(head_size) for _ in range(num_heads)])
        self.proj = nn.Linear(n_embd, n_embd)
        self.dropout = nn.Dropout(dropout)
    
    def forward(self, x):
        out = torch.cat([head(x) for head in self.heads], dim=-1)
        out = self.dropout(self.proj(out))
        return out

class FeedForward(nn.Module):
    """a simple layer followed by a non-linearity"""
    def __init__(self, n_embd):
        super().__init__()
        self.net = nn.Sequential(
            nn.Linear(n_embd, 4*n_embd),
            nn.ReLU(),
            nn.Linear(4*n_embd, n_embd),
            nn.Dropout(dropout)
        )

    def forward(self, x):
        return self.net(x)

class Block(nn.Module):
    """Transformer block: communication followed by computation"""
    def __init__(self, n_embd, n_head):
        super().__init__()
        head_size = n_embd // n_head
        self.sa_heads = MultiHeadAttention(n_head, head_size)
        self.ffwd = FeedForward(n_embd)
        self.ln1 = nn.LayerNorm(n_embd)
        self.ln2 = nn.LayerNorm(n_embd)

    def forward(self, x):
        x = x + self.sa_heads(self.ln1(x))
        x = x + self.ffwd(self.ln2(x))
        return x

class Model(nn.Module):
    def __init__(self, ):
        super().__init__()
        self.token_embedding_table = nn.Embedding(vocab_size, n_embd)
        self.position_embedding_table = nn.Embedding(block_size, n_embd)
        self.blocks = nn.Sequential(*[Block(n_embd, n_head) for _ in range(n_layer)])
        self.ln = nn.LayerNorm(n_embd) 
        self.lm_head = nn.Linear(n_embd, vocab_size)


    def forward(self, idx, targets=None):
        B, T = idx.shape

        # idx and targets are both (B,T) tensor of integers
        tok_emb = self.token_embedding_table(idx) # (B, T, C)
        pos_emb = self.position_embedding_table(torch.arange(T, device=device)) # (T, C)
        x = tok_emb + pos_emb # (B, T, C)
        x = self.blocks(x) # (B, T, C)
        x = self.ln(x) # (B, T, C)
        logits = self.lm_head(x) # (B, T, vocab_size)

        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
    
    # why would you ever want to take in a batch here, probably just want to generate one at a time right?
    def generate(self, idx=None, max_new_tokens=1):
        # idx is (B, T) array of indices in the current context 
        for _ in range(max_new_tokens):
            B, T = idx.shape
            # crop idx to last block_size tokens 
            idx_cond = idx[:, -block_size:] # (B, T)
            # get the predictions 
            logits, _loss = self(idx_cond)
            # focus only on the last time step 
            logits = logits[:, -1, :] # becomes (B, C)
            # apply softmax to get probabilities
            probs = F.softmax(logits, dim=-1) # (B, C)
            # sample from the distribution
            idx_next = torch.multinomial(probs, num_samples=1) # (B, 1)
            # append sampled index to the running sequence 
            idx = torch.cat((idx, idx_next), dim=1) # (B, T+1)
        return idx
    
    # generate from scratch, single batch
    def generate2(self, max_new_tokens):
        context = torch.zeros(1, dtype=torch.long, device=device)
        for _ in range(max_new_tokens):
            # get the current length of the context
            T = context.shape[0] 
            # crop the context to last block_size tokens  
            cropped_context = context[-block_size:] 
            # get the predictions (unsqueeze to add dummy batch dimension since generate expects a batch)
            logits, _loss = self(torch.unsqueeze(cropped_context, dim=0))
            # squeeze out the dummy batch dimension
            logits = logits.squeeze(0)
            # focus only on the last time step
            logits = logits[-1, :]
            # apply softmax to get probabilities
            probs = F.softmax(logits, dim=-1)
            # sample from the distribution
            idx_next = torch.multinomial(probs, num_samples=1)
            # append sampled index to the running sequence
            context = torch.cat((context, idx_next), dim=0)
        return context.tolist()

    
    
model = Model()
m = model.to(device)

# create a pytorch optimizer 
optimizer = torch.optim.AdamW(model.parameters(), lr=learning_rate)

start = time.time()
for iter in range(max_iters):
    # sample a batch of data 
    xb, yb = get_batch('train')

    # evaluate the loss 
    logits, loss = model(xb, yb)
    optimizer.zero_grad(set_to_none=True)
    loss.backward()
    optimizer.step()

    # every once in a while evaluate the loss on trai/val 
    if iter % eval_interval == 0:
        losses = estimate_loss()
        print(f"step {iter}/{max_iters}: train loss {losses['train']:.4f}, val loss {losses['val']:.4f}")

# print time elapsed
end = time.time()
print(f"total training time {humanize.naturaldelta(end - start)}")

# generate from the model 
print(decode(m.generate2(max_new_tokens=500)))