File size: 11,836 Bytes
7fc0f78 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 |
from dataclasses import dataclass
import torch
import torch.nn as nn
import torch.nn.functional as F
from model import GPT, GPTConfig
import tiktoken
from torch.utils.data import Dataset, DataLoader, DistributedSampler
import math
import matplotlib.pyplot as plt
from torch.distributed import init_process_group, destroy_process_group
from torch.nn.parallel import DistributedDataParallel as DDP
import torch.distributed as dist
import os
import signal
import sys
import numpy as np
import time
import logging
def seconds_to_hms(seconds):
return time.strftime('%H:%M:%S', time.gmtime(seconds))
def signal_handler(sig, frame):
print('Gracefully stopping the training process')
destroy_process_group()
sys.exit(0)
signal.signal(signal.SIGINT, signal_handler)
manual_seed = 1339
torch.manual_seed(manual_seed)
if torch.cuda.is_available():
torch.cuda.manual_seed(manual_seed)
# ***************************#
# Device Configuration
# ***************************#
device = torch.device("cpu")
if torch.cuda.is_available():
device = torch.device("cuda")
elif hasattr(torch.backends, "mps") and torch.backends.mps.is_available():
device = torch.device("mps")
print("Using device:", device)
# ***************************#
# Tokenizer Setup
# ***************************#
enc = tiktoken.get_encoding('gpt2')
lossi = []
val_lossi = []
# ***************************#
# Load Text Data
# ***************************#
with open("tinyshakespeare.txt", "r") as f:
text = f.read()
tokens = enc.encode(text)
print(f"Number of tokens: {len(tokens):,}")
# ***************************#
# Set up DDP
# ***************************#
# torchrun command sets the env variables RANK, LOCAL_RANK, and WORLD_SIZE
ddp = int(os.environ.get('RANK', -1)) != -1 # is this a ddp run?
if ddp:
# use of DDP atm demands CUDA, we set the device appropriately according to rank
assert torch.cuda.is_available(), "for now i think we need CUDA for DDP"
init_process_group(backend='nccl')
ddp_rank = int(os.environ['RANK'])
ddp_local_rank = int(os.environ['LOCAL_RANK'])
ddp_world_size = int(os.environ['WORLD_SIZE'])
device = f'cuda:{ddp_local_rank}'
torch.cuda.set_device(device)
# this process will do logging, checkpointing etc.
master_process = ddp_rank == 0
else:
# vanilla, non-DDP run
ddp_rank = 0
ddp_local_rank = 0
ddp_world_size = 1
master_process = True
if master_process:
print(f"ddp: {ddp}, rank: {ddp_rank}, local_rank: {ddp_local_rank}, world_size: {ddp_world_size}, master_process: {master_process}")
# ***************************#
# Model Configuration
# ***************************#
gpt = GPT(GPTConfig(vocab_size=50304), master_process).to(device)
if device == torch.device("cuda"):
gpt.compile()
if ddp:
gpt = DDP(gpt, device_ids=[ddp_local_rank])
raw_gpt = gpt.module if ddp else gpt
# ***************************#
# Dataset and Dataloader
# ***************************#
def load_tokens(filename):
npt = np.load(filename)
npt = npt.astype(np.int32) # added after video
ptt = torch.tensor(npt, dtype=torch.long)
return ptt
class DataLoader_Custom:
def __init__(self, B, T, process_rank, num_processes, split, shuffle=False):
self.B = B
self.T = T
self.process_rank = process_rank
self.num_processes = num_processes
self.shuffle = shuffle
assert split in ["train", "val"]
data_root = "edu_fineweb10B"
shards = os.listdir(data_root)
shards = [s for s in shards if split in s]
shards = sorted(shards)
shards = [os.path.join(data_root, s) for s in shards]
self.shards = shards
assert len(shards) > 0, "No shards found for split {}".format(split)
if master_process:
print("Found {} shards for split {}".format(len(shards), split))
self.current_shard = 0
self.tokens = load_tokens(self.shards[self.current_shard])
self.current_position = self.B * self.T * self.process_rank
def next_batch(self):
B, T = self.B, self.T
buf = self.tokens[self.current_position:self.current_position + B*T+1]
x = buf[:-1].view(B, T)
y = buf[1:].view(B, T)
self.current_position += B*T * self.num_processes
if self.current_position + (B*T*self.num_processes+1) > len(self.tokens):
self.current_shard = self.current_shard + 1 % len(self.shards)
self.tokens = load_tokens(self.shards[self.current_shard])
self.current_position = self.B * self.T * self.process_rank
return x, y
def reset(self):
self.current_shard = 0
self.tokens = load_tokens(self.shards[self.current_shard])
self.current_position = self.B * self.T * self.process_rank
T = 4
batch_size = 1
total_batch_size = 2**2 # 524,288 = 2**19, in number of tokens
assert total_batch_size % (T*batch_size*ddp_world_size) == 0, "Batch size is not divisible by B*T"
grad_accum_steps = total_batch_size // (T*batch_size*ddp_world_size)
if master_process:
print("Total desired batch size: {:,}".format(total_batch_size))
print("gradient accumulation steps: {:,}".format(grad_accum_steps))
train_dataloader = DataLoader_Custom(batch_size, T, ddp_local_rank, ddp_world_size, "train")
val_dataloader = DataLoader_Custom(batch_size, T, ddp_local_rank, ddp_world_size, "val")
# ***************************#
# Text Generation Function
# ***************************#
def generate_text(seed_text, model, enc, max_len=100, print_while_generating=True):
if print_while_generating:
print(seed_text, end="")
model.eval()
with torch.no_grad():
tokens = enc.encode(seed_text)
for _ in range(max_len):
x = torch.tensor(tokens[-T:], dtype=torch.long,
device=device).unsqueeze(0)
logits, _ = model(x)
next_token = torch.argmax(logits[:, -1, :])
tokens.append(int(next_token))
if print_while_generating:
print(enc.decode([int(next_token)]), end="")
print()
return enc.decode(tokens)
# ***************************#
# Optimizer Configuration
# ***************************#
if ddp:
optimizer = raw_gpt.configure_optimizers(
weight_decay=0.1, learning_rate=6e-4, device=device)
else:
optimizer = gpt.configure_optimizers(
weight_decay=0.1, learning_rate=6e-4, device=device)
torch.set_float32_matmul_precision('high')
# ***************************#
# Learning Rate Scheduler
# ***************************#
max_lr = 6e-4
min_lr = max_lr * 0.1
warmup_steps = 715
max_steps = 50
def get_lr(step):
if step < warmup_steps:
return max_lr * (step+1) / warmup_steps
if step > max_steps:
return min_lr
decay_ratio = (step - warmup_steps) / (max_steps - warmup_steps)
assert 0 <= decay_ratio <= 1
coeff = 0.5 * (1.0 + math.cos(math.pi * decay_ratio))
return min_lr + coeff * (max_lr - min_lr)
# Check if the device supports bfloat16
supports_bfloat16 = False
if device == "cuda":
capability = torch.cuda.get_device_capability()
if capability[0] >= 8 and capability[1] >= 0:
supports_bfloat16 = True
print("Supports bfloat16:", supports_bfloat16)
# ***************************#
# Training Loop
# ***************************#
generate_every = 50
validate_every = 10
save_every = 5
t0 = time.time()
# Initialize logging
logging.basicConfig(level=logging.INFO, format='%(message)s')
logger = logging.getLogger(__name__)
# Add a file handler
file_handler = logging.FileHandler('training_log.txt')
file_handler.setLevel(logging.INFO)
file_handler.setFormatter(logging.Formatter('%(message)s'))
logger.addHandler(file_handler)
for step in range(max_steps):
loss_accum = 0.0
gpt.zero_grad()
for minibatchstep in range(grad_accum_steps):
x, y = train_dataloader.next_batch()
x, y = x.to(device), y.to(device)
if supports_bfloat16:
with torch.autocast(device_type='cuda', dtype=torch.bfloat16):
logits, loss = gpt(x, y)
else:
logits, loss = gpt(x, y)
loss = loss / grad_accum_steps
loss_accum += loss.detach()
if ddp:
gpt.require_backward_grad_sync = (minibatchstep == grad_accum_steps - 1)
loss.backward()
if ddp:
dist.all_reduce(loss_accum, op=dist.ReduceOp.AVG)
lossi.append(loss_accum.item())
norm = torch.nn.utils.clip_grad_norm_(gpt.parameters(), 1.0)
lr = get_lr(step)
for param_group in optimizer.param_groups:
param_group['lr'] = lr
optimizer.step()
t_current = time.time()
elapsed_time = t_current - t0
steps_completed = step + 1
avg_time_per_step = elapsed_time / steps_completed
remaining_steps = max_steps - steps_completed
remaining_time = remaining_steps * avg_time_per_step
if master_process:
logger.info(f'Step {step} | Loss: {loss_accum:.6f} | Norm: {norm:.4f} | LR: {lr:.2e} | Time: {seconds_to_hms(elapsed_time)} | Remaining: {seconds_to_hms(remaining_time)} | Avg Time/Step: {avg_time_per_step:.2f}')
if master_process and step % generate_every == 0:
generated_text = generate_text("The king said", gpt, enc, max_len=25, print_while_generating=False)
logger.info(f'Generated Text at Step {step}: {generated_text}')
# Validation step
if step % validate_every == 0:
if master_process:
logger.info("Validating...")
gpt.eval()
val_loss_accum = 0.0
val_dataloader.reset()
with torch.no_grad():
val_loss_accum
val_loss_steps = 20
for _ in range(val_loss_steps):
x, y = val_dataloader.next_batch()
x, y = x.to(device), y.to(device)
if supports_bfloat16:
with torch.autocast(device_type='cuda', dtype=torch.bfloat16):
val_logits, val_loss = gpt(x, y)
else:
val_logits, val_loss = gpt(x, y)
val_loss = val_loss / val_loss_steps
val_loss_accum += val_loss.detach()
if ddp:
dist.all_reduce(val_loss_accum, op=dist.ReduceOp.AVG)
if master_process:
logger.info(f'Validation Loss: {val_loss_accum}')
val_lossi.append(val_loss_accum.item())
if step % save_every == 0 and master_process:
print("Saving model and loss...")
torch.save(raw_gpt.state_dict(), "gpt2_step_{}.pth".format(step))
torch.save(torch.tensor(lossi), "lossi_step_{}.pth".format(step))
torch.save(torch.tensor(val_lossi), "val_lossi_step_{}.pth".format(step))
# ***************************#
# Plot Loss
# ***************************#
plot = True
if master_process and plot:
plt.plot(lossi, label="Train Loss")
# Stretch val_lossi to match the length of lossi
val_lossi_stretched = np.interp(
np.linspace(0, len(val_lossi) - 1, len(lossi)),
np.arange(len(val_lossi)),
val_lossi
)
plt.plot(val_lossi_stretched, label="Validation Loss")
plt.legend()
plt.xlabel("Step")
plt.ylabel("Loss")
plt.show()
# Generate Final Text
if master_process:
print(generate_text("The king said", gpt, enc, max_len=25, print_while_generating=False))
# ***************************#
# Save Model and Loss
# ***************************#
if master_process:
torch.save(gpt.state_dict(), "gpt2_shakespeare.pth")
torch.save(torch.tensor(lossi), "lossi.pth")
torch.save(torch.tensor(val_lossi), "val_lossi.pth")
# ***************************#
# Cleanup
# ***************************#
if ddp:
destroy_process_group()
import sys; sys.exit(0)
|