File size: 40,122 Bytes
b9049d2 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 | #!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
train_nlp_h100_optimized.py โ v2 (bugfix device mismatch)
===========================================================
Corrections vs v1 :
โข apply_qlora() appelรฉ APRรS model.to(device) โ lora_A/lora_B naissent sur CUDA
โข LoRALinear.__init__ : move explicite des adaptateurs sur le device du base_layer
โข torch.compile dรฉsactivรฉ quand USE_CHECKPOINTING=True (conflict dynamo+checkpoint
avec sous-modules custom) โ on utilise COMPILE_AFTER_CKPT pour les cas oรน on
veut quand mรชme compiler (USE_CHECKPOINTING=False)
โข Ajout d'un fallback propre : si compile crash, on continue sans compile
"""
from __future__ import annotations
import itertools
import json
import math
import os
import random
import time
from collections import OrderedDict
from contextlib import nullcontext
from dataclasses import asdict, dataclass
from pathlib import Path
from typing import Iterator, Optional
import torch
import torch.distributed as dist
import torch.nn as nn
import torch.nn.functional as F
try:
import bitsandbytes as bnb
from bitsandbytes.nn import Params4bit
HAS_BNB = True
except ImportError:
HAS_BNB = False
print("[warn] bitsandbytes non disponible โ quantification 4-bit dรฉsactivรฉe")
try:
from flash_attn import flash_attn_func
HAS_FLASH = True
except ImportError:
HAS_FLASH = False
print("[warn] flash-attn non disponible โ fallback F.scaled_dot_product_attention")
from datasets import load_dataset
from torch.nn.parallel import DistributedDataParallel as DDP
from tokenizers import (
Tokenizer, decoders, models, normalizers,
pre_tokenizers, processors, trainers,
)
from transformers import PreTrainedTokenizerFast
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ CHEMINS โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
OUT_DIR = Path("./nlp_1b_h100_opt")
OUT_DIR.mkdir(parents=True, exist_ok=True)
TOKENIZER_DIR = OUT_DIR / "tokenizer_32k"
CONFIG_FILE = OUT_DIR / "config.json"
MODEL_FILE = OUT_DIR / "model.pt"
BEST_MODEL_FILE= OUT_DIR / "model_best.pt"
STATE_FILE = OUT_DIR / "train_state.pt"
BASE_CHECKPOINT: Optional[Path] = None
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ HYPERPARAMรTRES โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
SEED = 42
TARGET_VRAM_GIB= 78.0
BLOCK_SIZE = 1024
VOCAB_SIZE = 32_000
D_MODEL = 1536
N_HEADS = 24
N_LAYERS = 24
D_FF = 6144
DROPOUT = 0.0
USE_QLORA = True
LORA_R = 64
LORA_ALPHA = 128
LORA_DROPOUT = 0.05
LORA_TARGET_MODULES = ["qkv", "proj", "w1", "w2", "w3"]
NUM_EPOCHS = 10
LEARNING_RATE = 3e-4
MIN_LR = 3e-5
WEIGHT_DECAY = 0.1
WARMUP_STEPS = 500
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ RรGLAGE BATCH SIZE โ 78 Go VRAM โ
# โ Dรฉmarrer : BATCH_SIZE=8, GRAD_ACCUM_STEPS=2 โ
# โ Augmenter BATCH_SIZE par +2 jusqu'ร max_reserved โ 77 Go dans les logs โ
# โ Si OOM : BATCH_SIZE -= 1 ou USE_CHECKPOINTING=True โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
BATCH_SIZE = 16
GRAD_ACCUM_STEPS = 1
MAX_GRAD_NORM = 1.0
EVAL_EVERY = 500
SAVE_EVERY = 1_000
DTYPE = torch.bfloat16
# โโ Compile : dรฉsactivรฉ quand USE_CHECKPOINTING=True pour รฉviter le conflict
# dynamo โ checkpoint โ sous-modules custom (LoRALinear).
# Mettre USE_CHECKPOINTING=False ET USE_COMPILE=True pour vitesse max.
USE_CHECKPOINTING = False # รฉconomise ~8ร activations VRAM
USE_COMPILE = True # โ mettre True seulement si USE_CHECKPOINTING=False
COMPILE_MODE = "reduce-overhead"
TRAIN_NUM_WORKERS = 4
EVAL_NUM_WORKERS = 2
PREFETCH_FACTOR = 2
TOKENIZER_SAMPLE_DOCS_PER_SOURCE = 15_000
TOKENIZER_CHAR_LIMIT = 2_000
TEXT_CHAR_LIMIT = 4_000
SPECIAL_TOKENS = ["<pad>", "<bos>", "<eos>", "<unk>"]
PAD_TOKEN, BOS_TOKEN, EOS_TOKEN, UNK_TOKEN = SPECIAL_TOKENS
WIKI_CONFIGS = ["20231101.en", "20231101.fr", "20231101.ar"]
FINEWEB_CONFIG = "sample-10BT"
DEV_DOCS_PER_WIKI_CONFIG = 1_500
DEV_DOCS_FINEWEB = 3_000
TRAIN_DOCS_PER_WIKI_CONFIG = 30_000
TRAIN_DOCS_FINEWEB = 60_000
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ DISTRIBUTED โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
def is_distributed() -> bool:
return dist.is_available() and dist.is_initialized()
def get_rank() -> int:
return dist.get_rank() if is_distributed() else 0
def get_world_size() -> int:
return dist.get_world_size() if is_distributed() else 1
def is_main() -> bool:
return get_rank() == 0
def init_distributed() -> Optional[torch.device]:
local_rank = int(os.environ.get("LOCAL_RANK", -1))
if local_rank == -1:
return None
dist.init_process_group("nccl")
torch.cuda.set_device(local_rank)
return torch.device(f"cuda:{local_rank}")
def set_seed(seed: int) -> None:
random.seed(seed)
torch.manual_seed(seed)
if torch.cuda.is_available():
torch.cuda.manual_seed_all(seed)
def get_device(ddp_device: Optional[torch.device] = None) -> torch.device:
if ddp_device is not None:
return ddp_device
if torch.cuda.is_available():
return torch.device(f"cuda:{torch.cuda.current_device()}")
return torch.device("cpu")
def current_cuda_index(device: torch.device) -> int:
return device.index if device.index is not None else torch.cuda.current_device()
def autocast_context(device: torch.device):
if device.type == "cuda":
return torch.autocast("cuda", dtype=DTYPE)
return nullcontext()
def unwrap_model(model: nn.Module) -> nn.Module:
m = model.module if isinstance(model, DDP) else model
return m._orig_mod if hasattr(m, "_orig_mod") else m
def count_parameters(model: nn.Module, trainable_only: bool = True) -> int:
return sum(p.numel() for p in model.parameters() if not trainable_only or p.requires_grad)
def normalize_state_dict_keys(sd: dict) -> OrderedDict:
out = OrderedDict()
for k, v in sd.items():
for prefix in ("module._orig_mod.", "_orig_mod.", "module."):
if k.startswith(prefix):
k = k[len(prefix):]
break
out[k] = v
return out
def normalize_text(t: str) -> str:
return " ".join(t.strip().split())
def safe_str(x) -> str:
return x if isinstance(x, str) else ("" if x is None else str(x))
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ DATASETS โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
def load_wiki_stream(cfg_name: str):
return load_dataset("wikimedia/wikipedia", cfg_name, split="train", streaming=True)
def load_fineweb_stream():
return load_dataset("HuggingFaceFW/fineweb-edu", FINEWEB_CONFIG, split="train", streaming=True)
def stream_texts(ds, start: int, count: int, char_limit: int) -> Iterator[str]:
for row in itertools.islice(ds, start, start + count):
text = normalize_text(safe_str(row.get("text", "")))
if len(text) >= 20:
yield text[:char_limit]
def tokenizer_training_iterator() -> Iterator[str]:
for c in WIKI_CONFIGS:
yield from stream_texts(load_wiki_stream(c), 0, TOKENIZER_SAMPLE_DOCS_PER_SOURCE, TOKENIZER_CHAR_LIMIT)
yield from stream_texts(load_fineweb_stream(), 0, TOKENIZER_SAMPLE_DOCS_PER_SOURCE, TOKENIZER_CHAR_LIMIT)
def build_epoch_train_texts(epoch: int) -> list[str]:
texts: list[str] = []
for c in WIKI_CONFIGS:
start = DEV_DOCS_PER_WIKI_CONFIG + epoch * TRAIN_DOCS_PER_WIKI_CONFIG
texts.extend(stream_texts(load_wiki_stream(c), start, TRAIN_DOCS_PER_WIKI_CONFIG, TEXT_CHAR_LIMIT))
start = DEV_DOCS_FINEWEB + epoch * TRAIN_DOCS_FINEWEB
texts.extend(stream_texts(load_fineweb_stream(), start, TRAIN_DOCS_FINEWEB, TEXT_CHAR_LIMIT))
random.Random(SEED + epoch).shuffle(texts)
return texts
def build_eval_texts() -> list[str]:
texts: list[str] = []
for c in WIKI_CONFIGS:
texts.extend(stream_texts(load_wiki_stream(c), 0, DEV_DOCS_PER_WIKI_CONFIG, TEXT_CHAR_LIMIT))
texts.extend(stream_texts(load_fineweb_stream(), 0, DEV_DOCS_FINEWEB, TEXT_CHAR_LIMIT))
return texts
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ PACKED DATASET โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
class PackedTextList(torch.utils.data.IterableDataset):
def __init__(self, texts, tokenizer, block_size, epoch_seed=0):
super().__init__()
self.texts = texts
self.tokenizer = tokenizer
self.block_size = block_size
self.epoch_seed = epoch_seed
def __iter__(self):
worker = torch.utils.data.get_worker_info()
rank, ws = get_rank(), get_world_size()
if worker is None:
shard_mod, shard_id = ws, rank
else:
shard_mod = worker.num_workers * ws
shard_id = rank * worker.num_workers + worker.id
rng = random.Random(self.epoch_seed)
indices = list(range(len(self.texts)))
rng.shuffle(indices)
bos, eos = self.tokenizer.bos_token_id, self.tokenizer.eos_token_id
buf: list[int] = []
for li, ti in enumerate(indices):
if li % shard_mod != shard_id:
continue
ids = self.tokenizer.encode(self.texts[ti], add_special_tokens=False)
if not ids:
continue
buf.extend([bos] + ids + [eos])
while len(buf) >= self.block_size + 1:
chunk = buf[: self.block_size + 1]
buf = buf[self.block_size + 1 :]
yield {
"input_ids": torch.tensor(chunk[:-1], dtype=torch.long),
"labels": torch.tensor(chunk[1:], dtype=torch.long),
}
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ TOKENIZER โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
def tokenizer_ready() -> bool:
return (TOKENIZER_DIR / "tokenizer.json").exists() and (TOKENIZER_DIR / "tokenizer_config.json").exists()
def train_tokenizer_once() -> None:
TOKENIZER_DIR.mkdir(parents=True, exist_ok=True)
tok = Tokenizer(models.BPE(unk_token=UNK_TOKEN))
tok.normalizer = normalizers.Sequence([normalizers.NFKC()])
tok.pre_tokenizer = pre_tokenizers.ByteLevel(add_prefix_space=False)
tok.decoder = decoders.ByteLevel()
trainer = trainers.BpeTrainer(
vocab_size=VOCAB_SIZE, min_frequency=2, show_progress=is_main(),
special_tokens=SPECIAL_TOKENS, initial_alphabet=pre_tokenizers.ByteLevel.alphabet(),
)
tok.train_from_iterator(tokenizer_training_iterator(), trainer=trainer)
bos_id, eos_id = tok.token_to_id(BOS_TOKEN), tok.token_to_id(EOS_TOKEN)
tok.post_processor = processors.TemplateProcessing(
single=f"{BOS_TOKEN} $A {EOS_TOKEN}",
pair=f"{BOS_TOKEN} $A {EOS_TOKEN} $B:1 {EOS_TOKEN}:1",
special_tokens=[(BOS_TOKEN, bos_id), (EOS_TOKEN, eos_id)],
)
tok.save(str(TOKENIZER_DIR / "tokenizer.json"))
fast = PreTrainedTokenizerFast(
tokenizer_file=str(TOKENIZER_DIR / "tokenizer.json"),
bos_token=BOS_TOKEN, eos_token=EOS_TOKEN, unk_token=UNK_TOKEN, pad_token=PAD_TOKEN,
)
fast.save_pretrained(str(TOKENIZER_DIR))
def train_or_load_tokenizer() -> PreTrainedTokenizerFast:
TOKENIZER_DIR.mkdir(parents=True, exist_ok=True)
if not tokenizer_ready():
if is_distributed():
if is_main():
print("Entraรฎnement tokenizer 32kโฆ"); train_tokenizer_once()
dist.barrier()
else:
print("Entraรฎnement tokenizer 32kโฆ"); train_tokenizer_once()
return PreTrainedTokenizerFast.from_pretrained(str(TOKENIZER_DIR))
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ MODรLE โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
@dataclass
class GPTConfig:
vocab_size: int = VOCAB_SIZE
block_size: int = BLOCK_SIZE
d_model: int = D_MODEL
n_heads: int = N_HEADS
n_layers: int = N_LAYERS
d_ff: int = D_FF
dropout: float = DROPOUT
use_checkpointing: bool = USE_CHECKPOINTING
class RMSNorm(nn.Module):
def __init__(self, dim: int, eps: float = 1e-6):
super().__init__()
self.weight = nn.Parameter(torch.ones(dim))
self.eps = eps
def forward(self, x):
return self.weight * x * torch.rsqrt(x.pow(2).mean(-1, keepdim=True) + self.eps)
class RotaryEmbedding(nn.Module):
def __init__(self, dim: int, base: int = 10_000, max_seq: int = 4_096):
super().__init__()
inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2).float() / dim))
self.register_buffer("inv_freq", inv_freq, persistent=False)
t = torch.arange(max_seq).float()
freqs = torch.outer(t, inv_freq)
self.register_buffer("cos_cache", torch.repeat_interleave(freqs.cos(), 2, dim=-1), persistent=False)
self.register_buffer("sin_cache", torch.repeat_interleave(freqs.sin(), 2, dim=-1), persistent=False)
def forward(self, seq_len: int, dtype: torch.dtype):
return self.cos_cache[:seq_len].to(dtype), self.sin_cache[:seq_len].to(dtype)
def rotate_half(x):
x1, x2 = x[..., ::2], x[..., 1::2]
return torch.stack((-x2, x1), dim=-1).flatten(-2)
def apply_rope(x, cos, sin):
return x * cos.unsqueeze(0).unsqueeze(0) + rotate_half(x) * sin.unsqueeze(0).unsqueeze(0)
class CausalSelfAttention(nn.Module):
def __init__(self, cfg: GPTConfig):
super().__init__()
assert cfg.d_model % cfg.n_heads == 0
self.n_heads = cfg.n_heads
self.head_dim = cfg.d_model // cfg.n_heads
self.qkv = nn.Linear(cfg.d_model, 3 * cfg.d_model, bias=False)
self.proj = nn.Linear(cfg.d_model, cfg.d_model, bias=False)
self.dropout_p = cfg.dropout
self.rope = RotaryEmbedding(self.head_dim)
def forward(self, x):
b, t, c = x.shape
q, k, v = self.qkv(x).split(c, dim=-1)
q = q.view(b, t, self.n_heads, self.head_dim).transpose(1, 2)
k = k.view(b, t, self.n_heads, self.head_dim).transpose(1, 2)
v = v.view(b, t, self.n_heads, self.head_dim).transpose(1, 2)
cos, sin = self.rope(t, x.dtype)
q, k = apply_rope(q, cos, sin), apply_rope(k, cos, sin)
if HAS_FLASH:
# Flash Attention 2 attend (b, t, nh, hd)
q = q.transpose(1, 2)
k = k.transpose(1, 2)
v = v.transpose(1, 2)
y = flash_attn_func(q, k, v, dropout_p=self.dropout_p if self.training else 0.0, causal=True)
y = y.reshape(b, t, c)
else:
y = F.scaled_dot_product_attention(q, k, v, dropout_p=self.dropout_p if self.training else 0.0, is_causal=True)
y = y.transpose(1, 2).contiguous().view(b, t, c)
return self.proj(y)
class SwiGLU(nn.Module):
def __init__(self, cfg: GPTConfig):
super().__init__()
self.w1 = nn.Linear(cfg.d_model, cfg.d_ff, bias=False)
self.w2 = nn.Linear(cfg.d_model, cfg.d_ff, bias=False)
self.w3 = nn.Linear(cfg.d_ff, cfg.d_model, bias=False)
def forward(self, x):
return self.w3(F.silu(self.w1(x)) * self.w2(x))
class Block(nn.Module):
def __init__(self, cfg: GPTConfig):
super().__init__()
self.ln1 = RMSNorm(cfg.d_model)
self.attn = CausalSelfAttention(cfg)
self.ln2 = RMSNorm(cfg.d_model)
self.ff = SwiGLU(cfg)
def forward(self, x):
x = x + self.attn(self.ln1(x))
x = x + self.ff(self.ln2(x))
return x
class GPT(nn.Module):
def __init__(self, cfg: GPTConfig):
super().__init__()
self.cfg = cfg
self.tok_emb = nn.Embedding(cfg.vocab_size, cfg.d_model)
self.blocks = nn.ModuleList([Block(cfg) for _ in range(cfg.n_layers)])
self.ln_f = RMSNorm(cfg.d_model)
self.lm_head = nn.Linear(cfg.d_model, cfg.vocab_size, bias=False)
self.lm_head.weight = self.tok_emb.weight # weight tying
self.apply(self._init_weights)
@staticmethod
def _init_weights(m):
if isinstance(m, (nn.Linear, nn.Embedding)):
nn.init.normal_(m.weight, 0.0, 0.02)
if isinstance(m, nn.Linear) and m.bias is not None:
nn.init.zeros_(m.bias)
def forward(self, input_ids, labels=None):
x = self.tok_emb(input_ids)
for block in self.blocks:
if self.cfg.use_checkpointing and self.training:
x = torch.utils.checkpoint.checkpoint(block, x, use_reentrant=False)
else:
x = block(x)
logits = self.lm_head(self.ln_f(x))
loss = None
if labels is not None:
loss = F.cross_entropy(logits.reshape(-1, logits.size(-1)), labels.reshape(-1), ignore_index=-100)
return logits, loss
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ QLORA โ
# โ โ
# โ CORRECTIF CLร : apply_qlora() DOIT รชtre appelรฉ APRรS model.to(device). โ
# โ LoRALinear dรฉtecte automatiquement le device du base_layer et y crรฉe โ
# โ lora_A / lora_B directement, sans besoin de .to() sรฉparรฉ. โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
class LoRALinear(nn.Module):
"""
Adaptateur LoRA autour d'un nn.Linear existant.
IMPORTANT : les sous-modules lora_A et lora_B sont crรฉรฉs sur le MรME
device que base_layer.weight via le move explicite ci-dessous.
C'est la correction du bug 'cuda:0 vs cpu' de la v1.
"""
def __init__(self, base_layer: nn.Linear, r: int = LORA_R, alpha: int = LORA_ALPHA, dropout: float = LORA_DROPOUT):
super().__init__()
self.base = base_layer
self.r = r
self.scale = alpha / r
in_f, out_f = base_layer.in_features, base_layer.out_features
# โโ Dรฉtecter le device du base_layer โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# base_layer.weight peut รชtre un Params4bit (pas de .device direct)
try:
dev = next(base_layer.parameters()).device
except StopIteration:
dev = torch.device("cpu")
# Crรฉer les adaptateurs DIRECTEMENT sur le bon device
self.lora_A = nn.Linear(in_f, r, bias=False, device=dev)
self.lora_B = nn.Linear(r, out_f, bias=False, device=dev)
self.drop = nn.Dropout(dropout)
# Initialisation standard LoRA
nn.init.kaiming_uniform_(self.lora_A.weight, a=math.sqrt(5))
nn.init.zeros_(self.lora_B.weight)
# Geler les poids de base
for p in self.base.parameters():
p.requires_grad = False
def forward(self, x):
return self.base(x) + self.lora_B(self.lora_A(self.drop(x))) * self.scale
def apply_qlora(model: GPT, device: torch.device) -> GPT:
"""
Remplace les couches cibles par LoRALinear.
ร appeler IMPรRATIVEMENT aprรจs model.to(device).
"""
if not USE_QLORA:
return model
replaced = 0
# Collecter d'abord pour รฉviter de modifier le dict pendant l'itรฉration
targets = []
for name, module in model.named_modules():
parts = name.split(".")
if parts[-1] in LORA_TARGET_MODULES and isinstance(module, nn.Linear):
targets.append((name, module))
for name, module in targets:
parts = name.split(".")
parent = model
for part in parts[:-1]:
parent = getattr(parent, part)
lora_layer = LoRALinear(module)
setattr(parent, parts[-1], lora_layer)
replaced += 1
if is_main():
print(f"QLoRA : {replaced} couches remplacรฉes (device={device}, NF4={HAS_BNB})")
return model
def freeze_base_weights(model: GPT) -> None:
"""Seuls lora_A et lora_B restent entraรฎnables."""
for name, p in model.named_parameters():
p.requires_grad = ("lora_A" in name or "lora_B" in name)
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ OPTIMIZER โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
def build_optimizer(model: nn.Module) -> torch.optim.Optimizer:
decay, no_decay = [], []
for name, p in unwrap_model(model).named_parameters():
if not p.requires_grad:
continue
(decay if p.ndim >= 2 and "weight" in name else no_decay).append(p)
groups = [
{"params": decay, "weight_decay": WEIGHT_DECAY},
{"params": no_decay, "weight_decay": 0.0},
]
if HAS_BNB:
return bnb.optim.PagedAdamW8bit(groups, lr=LEARNING_RATE, betas=(0.9, 0.95), eps=1e-8)
return torch.optim.AdamW(groups, lr=LEARNING_RATE, betas=(0.9, 0.95), eps=1e-8, fused=torch.cuda.is_available())
def cosine_lr(step: int, total_steps: int) -> float:
if step < WARMUP_STEPS:
return LEARNING_RATE * step / max(1, WARMUP_STEPS)
p = min(1.0, (step - WARMUP_STEPS) / max(1, total_steps - WARMUP_STEPS))
return MIN_LR + 0.5 * (LEARNING_RATE - MIN_LR) * (1 + math.cos(math.pi * p))
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ CHECKPOINT โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
def save_checkpoint(model, optimizer, epoch, step, best_loss, path):
raw = unwrap_model(model)
torch.save({
"model": normalize_state_dict_keys(raw.state_dict()),
"optimizer": optimizer.state_dict(),
"epoch": epoch, "step": step, "best_loss": best_loss,
"config": asdict(raw.cfg),
}, path)
def maybe_load_base_checkpoint(model, device):
if BASE_CHECKPOINT is None or not Path(BASE_CHECKPOINT).exists():
return
ckpt = torch.load(BASE_CHECKPOINT, map_location=device)
unwrap_model(model).load_state_dict(normalize_state_dict_keys(ckpt["model"]), strict=False)
def load_resume_checkpoint(model, optimizer, path, device):
ckpt = torch.load(path, map_location=device)
unwrap_model(model).load_state_dict(normalize_state_dict_keys(ckpt["model"]), strict=True)
try:
optimizer.load_state_dict(ckpt["optimizer"])
except Exception as e:
print(f"[warn] Optimizer state non repris: {e}")
return int(ckpt.get("epoch", 0)), int(ckpt.get("step", 0)), float(ckpt.get("best_loss", 1e9))
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ รVALUATION โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
@torch.no_grad()
def evaluate(model, loader, device, max_batches=200) -> float:
model.eval()
losses = []
for i, batch in enumerate(loader):
if i >= max_batches:
break
inp = batch["input_ids"].to(device, non_blocking=True)
lbl = batch["labels"].to(device, non_blocking=True)
with autocast_context(device):
_, loss = model(inp, lbl)
losses.append(loss.item())
model.train()
return sum(losses) / max(1, len(losses))
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ DATALOADER โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
def make_loader(dataset, batch_size, num_workers, is_cuda):
kwargs = dict(batch_size=batch_size, num_workers=num_workers, pin_memory=is_cuda)
if num_workers > 0:
kwargs["persistent_workers"] = True
kwargs["prefetch_factor"] = PREFETCH_FACTOR
return torch.utils.data.DataLoader(dataset, **kwargs)
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# โ MAIN โ
# โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
def main() -> None:
ddp_device = init_distributed()
set_seed(SEED + get_rank())
device = get_device(ddp_device)
is_cuda = device.type == "cuda"
cuda_idx = None
vram_fraction = None
if is_cuda:
torch.backends.cuda.matmul.allow_tf32 = True
torch.backends.cudnn.allow_tf32 = True
torch.set_float32_matmul_precision("high")
cuda_idx = current_cuda_index(device)
_, total = torch.cuda.mem_get_info(cuda_idx)
vram_fraction = min(TARGET_VRAM_GIB * (1024**3) / total, 0.999)
torch.cuda.memory.set_per_process_memory_fraction(vram_fraction, device=cuda_idx)
if is_main():
print("=" * 72)
print(" GPT ~1B | H100 80 Go | QLoRA + BF16 + TF32 | v2 (device fix)")
print("=" * 72)
print(f"Device : {device} | World: {get_world_size()} GPU(s)")
print(f"Flash-2 : {HAS_FLASH} | BNB 4-bit: {HAS_BNB} | QLoRA: {USE_QLORA}")
print(f"Grad ckpt: {USE_CHECKPOINTING} | Compile: {USE_COMPILE} ({COMPILE_MODE})")
if is_cuda:
free, total = torch.cuda.mem_get_info(cuda_idx)
print(f"GPU : {torch.cuda.get_device_name(cuda_idx)}")
print(f"VRAM : {total/1024**3:.1f} GiB | libre: {free/1024**3:.1f} GiB")
tokenizer = train_or_load_tokenizer()
cfg = GPTConfig(vocab_size=len(tokenizer))
if is_main():
CONFIG_FILE.write_text(json.dumps(asdict(cfg), indent=2, ensure_ascii=False), encoding="utf-8")
# โโ 1. Crรฉer le modรจle โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
model = GPT(cfg).to(device)
# โโ 2. Appliquer QLoRA APRรS .to(device) โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
# C'est la correction principale : lora_A/lora_B sont crรฉรฉs sur CUDA
if USE_QLORA:
model = apply_qlora(model, device)
freeze_base_weights(model)
maybe_load_base_checkpoint(model, device)
# โโ 3. torch.compile (seulement si USE_CHECKPOINTING=False) โโโโโโโโโโโโโโ
# La combinaison compile + checkpoint + LoRALinear custom est instable
# avec torch.dynamo sur PyTorch 2.x. Choisir l'un ou l'autre.
if USE_COMPILE and not USE_CHECKPOINTING and hasattr(torch, "compile"):
try:
model = torch.compile(model, mode=COMPILE_MODE)
if is_main():
print(f"torch.compile activรฉ ({COMPILE_MODE})")
except Exception as e:
if is_main():
print(f"[warn] torch.compile รฉchouรฉ ({e}) โ poursuite sans compile")
# โโ 4. DDP โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
if is_distributed():
model = DDP(model, device_ids=[device.index])
optimizer = build_optimizer(model)
# โโ Datasets โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
eval_texts = build_eval_texts()
eval_ds = PackedTextList(eval_texts, tokenizer, cfg.block_size, SEED + 999)
eval_loader = make_loader(eval_ds, BATCH_SIZE, EVAL_NUM_WORKERS, is_cuda)
init_texts = build_epoch_train_texts(0)
steps_per_epoch = max(1, len(init_texts) // BATCH_SIZE)
total_steps_est = steps_per_epoch * NUM_EPOCHS
# โโ Reprise โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
start_epoch, start_step, best_eval = 0, 0, 1e9
if STATE_FILE.exists():
try:
if is_main(): print(f"Reprise depuis {STATE_FILE}")
start_epoch, start_step, best_eval = load_resume_checkpoint(model, optimizer, STATE_FILE, device)
except Exception as e:
if is_main():
bad = STATE_FILE.with_suffix(".corrupt.pt")
print(f"[warn] Checkpoint illisible: {e}")
try: STATE_FILE.rename(bad)
except Exception: pass
start_epoch, start_step, best_eval = 0, 0, 1e9
if is_main():
raw = unwrap_model(model)
n_total = count_parameters(raw, False)
n_train = count_parameters(raw, True)
print(f"Paramรจtres totaux : {n_total/1e9:.3f}B")
print(f"Paramรจtres entraรฎnรฉs : {n_train/1e6:.1f}M ({100*n_train/max(1,n_total):.2f}%)")
print(f"Batch size : {BATCH_SIZE} | Grad accum: {GRAD_ACCUM_STEPS} | Effective: {BATCH_SIZE*GRAD_ACCUM_STEPS}")
print(f"Steps estimรฉs: {total_steps_est} | Eval texts: {len(eval_texts)}")
print()
print("โโ Conseil VRAM โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ")
print(" Surveille 'max_reserved=XX GiB' ร step 50.")
print(" Augmente BATCH_SIZE par +2 jusqu'ร ~77 Go rรฉservรฉs.")
print(" Si OOM : BATCH_SIZE -= 1 ou USE_CHECKPOINTING=True.")
print("โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ")
# โโ Boucle d'entraรฎnement โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
model.train()
optimizer.zero_grad(set_to_none=True)
global_step = start_step
t0 = time.time()
log_loss_sum = 0.0
log_loss_count = 0
tokens_since_log = 0
last_log = time.time()
if is_cuda:
torch.cuda.reset_peak_memory_stats(cuda_idx)
for epoch in range(start_epoch, NUM_EPOCHS):
if is_main():
print(f"\n{'='*20} Epoch {epoch+1}/{NUM_EPOCHS} {'='*20}")
train_texts = build_epoch_train_texts(epoch)
train_ds = PackedTextList(train_texts, tokenizer, cfg.block_size, SEED + epoch)
train_loader = make_loader(train_ds, BATCH_SIZE, TRAIN_NUM_WORKERS, is_cuda)
for micro_step, batch in enumerate(train_loader):
inp = batch["input_ids"].to(device, non_blocking=True)
lbl = batch["labels"].to(device, non_blocking=True)
with autocast_context(device):
_, loss = model(inp, lbl)
(loss / GRAD_ACCUM_STEPS).backward()
log_loss_sum += loss.item()
log_loss_count += 1
tokens_since_log += inp.numel()
if (micro_step + 1) % GRAD_ACCUM_STEPS != 0:
continue
lr = cosine_lr(global_step, total_steps_est)
for group in optimizer.param_groups:
group["lr"] = lr
torch.nn.utils.clip_grad_norm_(model.parameters(), MAX_GRAD_NORM)
optimizer.step()
optimizer.zero_grad(set_to_none=True)
global_step += 1
if global_step % 50 == 0 and is_main():
now = time.time()
elapsed = max(1e-6, now - last_log)
tok_s = tokens_since_log / elapsed
avg_loss = log_loss_sum / max(1, log_loss_count)
print(
f"ep {epoch+1}/{NUM_EPOCHS} | step={global_step:5d} | "
f"loss={avg_loss:.4f} | lr={lr:.2e} | {tok_s:,.0f} tok/s"
)
if is_cuda:
alloc = torch.cuda.memory_allocated(cuda_idx) / 1024**3
reserved = torch.cuda.memory_reserved(cuda_idx) / 1024**3
max_alloc = torch.cuda.max_memory_allocated(cuda_idx) / 1024**3
max_res = torch.cuda.max_memory_reserved(cuda_idx) / 1024**3
print(
f"GPU mem | alloc={alloc:.2f} | reserved={reserved:.2f} | "
f"max_alloc={max_alloc:.2f} | max_reserved={max_res:.2f} (GiB)"
)
last_log = now
tokens_since_log = 0
log_loss_sum = 0.0
log_loss_count = 0
if global_step % EVAL_EVERY == 0 and is_main():
val_loss = evaluate(model, eval_loader, device)
print(f"[eval] step {global_step:5d} | val_loss={val_loss:.4f}")
if val_loss < best_eval:
best_eval = val_loss
save_checkpoint(model, optimizer, epoch, global_step, best_eval, BEST_MODEL_FILE)
print(f"โ Meilleur modรจle โ {BEST_MODEL_FILE}")
if global_step % SAVE_EVERY == 0 and is_main():
save_checkpoint(model, optimizer, epoch, global_step, best_eval, STATE_FILE)
save_checkpoint(model, optimizer, epoch, global_step, best_eval, MODEL_FILE)
print(f"โ Checkpoint โ {MODEL_FILE}")
if is_main():
save_checkpoint(model, optimizer, epoch + 1, global_step, best_eval, STATE_FILE)
ckpt = OUT_DIR / f"model_epoch_{epoch+1:02d}.pt"
save_checkpoint(model, optimizer, epoch + 1, global_step, best_eval, ckpt)
print(f"โ Fin epoch {epoch+1}/{NUM_EPOCHS} โ {ckpt}")
if is_main():
save_checkpoint(model, optimizer, NUM_EPOCHS, global_step, best_eval, MODEL_FILE)
save_checkpoint(model, optimizer, NUM_EPOCHS, global_step, best_eval, STATE_FILE)
total_min = (time.time() - t0) / 60
print(f"\nModรจle final โ {MODEL_FILE}")
print(f"Meilleur modรจle โ {BEST_MODEL_FILE}")
print(f"Temps total : {total_min:.1f} min | Steps: {global_step}")
if is_distributed():
dist.destroy_process_group()
if __name__ == "__main__":
main() |