注意
点击 此处 下载完整的示例代码
使用流水线并行训练 Transformer 模型¶
作者: Pritam Damania
本教程演示了如何使用流水线并行在多个 GPU 上训练大型 Transformer 模型。本教程是 使用 nn.Transformer 和 TorchText 进行序列到序列建模 教程的扩展，并扩展了相同的模型以演示如何使用流水线并行来训练 Transformer 模型。
先决条件
定义模型¶
在本教程中，我们将把 Transformer 模型拆分到两个 GPU 上，并使用流水线并行来训练模型。该模型与 使用 nn.Transformer 和 TorchText 进行序列到序列建模 教程中使用的模型完全相同，但被拆分为两个阶段。大部分参数属于 nn.TransformerEncoder 层。 nn.TransformerEncoder 本身包含 nlayers 个 nn.TransformerEncoderLayer。因此，我们的重点是 nn.TransformerEncoder，我们将模型拆分，使得一半的 nn.TransformerEncoderLayer 在一个 GPU 上，另一半在另一个 GPU 上。为此，我们将 Encoder 和 Decoder 部分提取到单独的模块中，然后构建一个表示原始 Transformer 模块的 nn.Sequential。
import sys
import math
import torch
import torch.nn as nn
import torch.nn.functional as F
import tempfile
from torch.nn import TransformerEncoder, TransformerEncoderLayer
if sys.platform == 'win32':
print('Windows platform is not supported for pipeline parallelism')
sys.exit(0)
if torch.cuda.device_count() < 2:
print('Need at least two GPU devices for this tutorial')
sys.exit(0)
class Encoder(nn.Module):
def __init__(self, ntoken, ninp, dropout=0.5):
super(Encoder, self).__init__()
self.pos_encoder = PositionalEncoding(ninp, dropout)
self.encoder = nn.Embedding(ntoken, ninp)
self.ninp = ninp
self.init_weights()
def init_weights(self):
initrange = 0.1
self.encoder.weight.data.uniform_(-initrange, initrange)
def forward(self, src):
# Need (S, N) format for encoder.
src = src.t()
src = self.encoder(src) * math.sqrt(self.ninp)
return self.pos_encoder(src)
class Decoder(nn.Module):
def __init__(self, ntoken, ninp):
super(Decoder, self).__init__()
self.decoder = nn.Linear(ninp, ntoken)
self.init_weights()
def init_weights(self):
initrange = 0.1
self.decoder.bias.data.zero_()
self.decoder.weight.data.uniform_(-initrange, initrange)
def forward(self, inp):
# Need batch dimension first for output of pipeline.
return self.decoder(inp).permute(1, 0, 2)
PositionalEncoding 模块注入了一些关于序列中标记的相对或绝对位置的信息。位置编码与嵌入具有相同的维度，以便两者可以相加。在这里，我们使用不同频率的 sine 和 cosine 函数。
class PositionalEncoding(nn.Module):
def __init__(self, d_model, dropout=0.1, max_len=5000):
super(PositionalEncoding, self).__init__()
self.dropout = nn.Dropout(p=dropout)
pe = torch.zeros(max_len, d_model)
position = torch.arange(0, max_len, dtype=torch.float).unsqueeze(1)
div_term = torch.exp(torch.arange(0, d_model, 2).float() * (-math.log(10000.0) / d_model))
pe[:, 0::2] = torch.sin(position * div_term)
pe[:, 1::2] = torch.cos(position * div_term)
pe = pe.unsqueeze(0).transpose(0, 1)
self.register_buffer('pe', pe)
def forward(self, x):
x = x + self.pe[:x.size(0), :]
return self.dropout(x)
加载和批处理数据¶
训练过程使用来自 torchtext 的 Wikitext-2 数据集。要访问 torchtext 数据集，请按照 https://github.com/pytorch/data 上的说明安装 torchdata。
词汇表对象（vocab object）基于训练数据集构建，用于将标记（tokens）数值化成张量。从顺序数据开始，batchify() 函数将数据集排列成列，在数据被分成大小为 batch_size 的批次后，裁剪掉任何剩余的标记。例如，以字母表作为序列（总长度为 26）且批次大小为 4，我们将字母表分成 4 个长度为 6 的序列。
\[\begin{bmatrix} \text{A} & \text{B} & \text{C} & \ldots & \text{X} & \text{Y} & \text{Z} \end{bmatrix} \Rightarrow \begin{bmatrix} \begin{bmatrix}\text{A} \\ \text{B} \\ \text{C} \\ \text{D} \\ \text{E} \\ \text{F}\end{bmatrix} & \begin{bmatrix}\text{G} \\ \text{H} \\ \text{I} \\ \text{J} \\ \text{K} \\ \text{L}\end{bmatrix} & \begin{bmatrix}\text{M} \\ \text{N} \\ \text{O} \\ \text{P} \\ \text{Q} \\ \text{R}\end{bmatrix} & \begin{bmatrix}\text{S} \\ \text{T} \\ \text{U} \\ \text{V} \\ \text{W} \\ \text{X}\end{bmatrix} \end{bmatrix}\]
模型将这些列视为独立的，这意味着无法学习 G 和 F 之间的依赖关系，但这允许更有效的批处理。
import torch
from torchtext.datasets import WikiText2
from torchtext.data.utils import get_tokenizer
from torchtext.vocab import build_vocab_from_iterator
train_iter = WikiText2(split='train')
tokenizer = get_tokenizer('basic_english')
vocab = build_vocab_from_iterator(map(tokenizer, train_iter), specials=["<unk>"])
vocab.set_default_index(vocab["<unk>"])
def data_process(raw_text_iter):
data = [torch.tensor(vocab(tokenizer(item)), dtype=torch.long) for item in raw_text_iter]
return torch.cat(tuple(filter(lambda t: t.numel() > 0, data)))
train_iter, val_iter, test_iter = WikiText2()
train_data = data_process(train_iter)
val_data = data_process(val_iter)
test_data = data_process(test_iter)
device = torch.device("cuda")
def batchify(data, bsz):
# Divide the dataset into ``bsz`` parts.
nbatch = data.size(0) // bsz
# Trim off any extra elements that wouldn't cleanly fit (remainders).
data = data.narrow(0, 0, nbatch * bsz)
# Evenly divide the data across the ``bsz` batches.
data = data.view(bsz, -1).t().contiguous()
return data.to(device)
batch_size = 20
eval_batch_size = 10
train_data = batchify(train_data, batch_size)
val_data = batchify(val_data, eval_batch_size)
test_data = batchify(test_data, eval_batch_size)
生成输入和目标序列的函数¶
get_batch() 函数为 Transformer 模型生成输入和目标序列。它将源数据细分为长度为 bptt 的块。对于语言建模任务，模型需要以下单词作为 Target。例如，当 bptt 的值为 2 时，我们将获得以下两个变量，其中 i = 0
需要注意的是，这些块沿着维度 0，与 Transformer 模型中的 S 维度一致。批次维度 N 沿着维度 1。
bptt = 25
def get_batch(source, i):
seq_len = min(bptt, len(source) - 1 - i)
data = source[i:i+seq_len]
target = source[i+1:i+1+seq_len].view(-1)
# Need batch dimension first for pipeline parallelism.
return data.t(), target
模型规模和管道初始化¶
为了演示使用管道并行训练大型 Transformer 模型，我们相应地扩展了 Transformer 层。我们使用 4096 的嵌入维度，4096 的隐藏大小，16 个注意力头和 12 个总的 Transformer 层（nn.TransformerEncoderLayer）。这将创建一个具有 **约 14 亿** 个参数的模型。
我们需要初始化 RPC 框架，因为 Pipe 通过 RRef 依赖于 RPC 框架，这允许将来扩展到跨主机管道。由于我们使用单个进程来驱动多个 GPU，因此我们需要仅使用单个工作进程初始化 RPC 框架。
然后，管道在一个 GPU 上初始化 8 个 Transformer 层，在另一个 GPU 上初始化 8 个 Transformer 层。
注意
出于效率目的，我们确保传递给 Pipe 的 nn.Sequential 仅包含两个元素（对应于两个 GPU），这允许 Pipe 仅使用两个分区并避免任何跨分区开销。
ntokens = len(vocab) # the size of vocabulary
emsize = 4096 # embedding dimension
nhid = 4096 # the dimension of the feedforward network model in ``nn.TransformerEncoder``
nlayers = 12 # the number of ``nn.TransformerEncoderLayer`` in ``nn.TransformerEncoder``
nhead = 16 # the number of heads in the Multihead Attention models
dropout = 0.2 # the dropout value
from torch.distributed import rpc
tmpfile = tempfile.NamedTemporaryFile()
rpc.init_rpc(
name="worker",
rank=0,
world_size=1,
rpc_backend_options=rpc.TensorPipeRpcBackendOptions(
init_method="file://{}".format(tmpfile.name),
# Specifying _transports and _channels is a workaround and we no longer
# will have to specify _transports and _channels for PyTorch
# versions >= 1.8.1
_transports=["ibv", "uv"],
_channels=["cuda_ipc", "cuda_basic"],
)
)
num_gpus = 2
partition_len = ((nlayers - 1) // num_gpus) + 1
# Add encoder in the beginning.
tmp_list = [Encoder(ntokens, emsize, dropout).cuda(0)]
module_list = []
# Add all the necessary transformer blocks.
for i in range(nlayers):
transformer_block = TransformerEncoderLayer(emsize, nhead, nhid, dropout)
if i != 0 and i % (partition_len) == 0:
module_list.append(nn.Sequential(*tmp_list))
tmp_list = []
device = i // (partition_len)
tmp_list.append(transformer_block.to(device))
# Add decoder in the end.
tmp_list.append(Decoder(ntokens, emsize).cuda(num_gpus - 1))
module_list.append(nn.Sequential(*tmp_list))
from torch.distributed.pipeline.sync import Pipe
# Build the pipeline.
chunks = 8
model = Pipe(torch.nn.Sequential(*module_list), chunks = chunks)
def get_total_params(module: torch.nn.Module):
total_params = 0
for param in module.parameters():
total_params += param.numel()
return total_params
print ('Total parameters in model: {:,}'.format(get_total_params(model)))
Total parameters in model: 1,444,261,998
运行模型¶
应用 CrossEntropyLoss 来跟踪损失，并使用 SGD 实现随机梯度下降方法作为优化器。初始学习率设置为 5.0。应用 StepLR 来调整每个 epoch 的学习率。在训练过程中，我们使用 nn.utils.clip_grad_norm_ 函数将所有梯度一起缩放，以防止梯度爆炸。
criterion = nn.CrossEntropyLoss()
lr = 5.0 # learning rate
optimizer = torch.optim.SGD(model.parameters(), lr=lr)
scheduler = torch.optim.lr_scheduler.StepLR(optimizer, 1.0, gamma=0.95)
import time
def train():
model.train() # Turn on the train mode
total_loss = 0.
start_time = time.time()
ntokens = len(vocab)
# Train only for 50 batches to keep script execution time low.
nbatches = min(50 * bptt, train_data.size(0) - 1)
for batch, i in enumerate(range(0, nbatches, bptt)):
data, targets = get_batch(train_data, i)
optimizer.zero_grad()
# Since the Pipe is only within a single host and process the ``RRef``
# returned by forward method is local to this node and can simply
# retrieved via ``RRef.local_value()``.
output = model(data).local_value()
# Need to move targets to the device where the output of the
# pipeline resides.
loss = criterion(output.view(-1, ntokens), targets.cuda(1))
loss.backward()
torch.nn.utils.clip_grad_norm_(model.parameters(), 0.5)
optimizer.step()
total_loss += loss.item()
log_interval = 10
if batch % log_interval == 0 and batch > 0:
cur_loss = total_loss / log_interval
elapsed = time.time() - start_time
print('| epoch {:3d} | {:5d}/{:5d} batches | '
'lr {:02.2f} | ms/batch {:5.2f} | '
'loss {:5.2f} | ppl {:8.2f}'.format(
epoch, batch, nbatches // bptt, scheduler.get_lr()[0],
elapsed * 1000 / log_interval,
cur_loss, math.exp(cur_loss)))
total_loss = 0
start_time = time.time()
def evaluate(eval_model, data_source):
eval_model.eval() # Turn on the evaluation mode
total_loss = 0.
ntokens = len(vocab)
# Evaluate only for 50 batches to keep script execution time low.
nbatches = min(50 * bptt, data_source.size(0) - 1)
with torch.no_grad():
for i in range(0, nbatches, bptt):
data, targets = get_batch(data_source, i)
output = eval_model(data).local_value()
output_flat = output.view(-1, ntokens)
# Need to move targets to the device where the output of the
# pipeline resides.
total_loss += len(data) * criterion(output_flat, targets.cuda(1)).item()
return total_loss / (len(data_source) - 1)
循环遍历 epoch。如果验证损失是我们迄今为止看到的最佳损失，则保存模型。每个 epoch 后调整学习率。
best_val_loss = float("inf")
epochs = 3 # The number of epochs
best_model = None
for epoch in range(1, epochs + 1):
epoch_start_time = time.time()
train()
val_loss = evaluate(model, val_data)
print('-' * 89)
print('| end of epoch {:3d} | time: {:5.2f}s | valid loss {:5.2f} | '
'valid ppl {:8.2f}'.format(epoch, (time.time() - epoch_start_time),
val_loss, math.exp(val_loss)))
print('-' * 89)
if val_loss < best_val_loss:
best_val_loss = val_loss
best_model = model
scheduler.step()
/opt/conda/envs/py_3.10/lib/python3.10/site-packages/torch/optim/lr_scheduler.py:402: UserWarning:
To get the last learning rate computed by the scheduler, please use `get_last_lr()`.
| epoch 1 | 10/ 50 batches | lr 5.00 | ms/batch 2937.68 | loss 51.97 | ppl 37278238304344674926592.00
| epoch 1 | 20/ 50 batches | lr 5.00 | ms/batch 2614.96 | loss 39.16 | ppl 101468412802272112.00
| epoch 1 | 30/ 50 batches | lr 5.00 | ms/batch 2618.78 | loss 45.74 | ppl 73373605537851539456.00
| epoch 1 | 40/ 50 batches | lr 5.00 | ms/batch 2620.70 | loss 39.05 | ppl 90831844662671120.00
-----------------------------------------------------------------------------------------
| end of epoch 1 | time: 148.36s | valid loss 1.59 | valid ppl 4.92
-----------------------------------------------------------------------------------------
| epoch 2 | 10/ 50 batches | lr 4.51 | ms/batch 2886.67 | loss 38.92 | ppl 79792098193225456.00
| epoch 2 | 20/ 50 batches | lr 4.51 | ms/batch 2625.91 | loss 33.86 | ppl 508484255367480.44
| epoch 2 | 30/ 50 batches | lr 4.51 | ms/batch 2628.64 | loss 29.47 | ppl 6267626426289.98
| epoch 2 | 40/ 50 batches | lr 4.51 | ms/batch 2629.61 | loss 20.07 | ppl 521065165.54
-----------------------------------------------------------------------------------------
| end of epoch 2 | time: 148.22s | valid loss 0.54 | valid ppl 1.71
-----------------------------------------------------------------------------------------
| epoch 3 | 10/ 50 batches | lr 4.29 | ms/batch 2889.23 | loss 13.75 | ppl 935925.21
| epoch 3 | 20/ 50 batches | lr 4.29 | ms/batch 2629.26 | loss 10.74 | ppl 46322.74
| epoch 3 | 30/ 50 batches | lr 4.29 | ms/batch 2628.18 | loss 10.97 | ppl 58152.80
| epoch 3 | 40/ 50 batches | lr 4.29 | ms/batch 2626.75 | loss 11.29 | ppl 80130.60
-----------------------------------------------------------------------------------------
| end of epoch 3 | time: 148.23s | valid loss 0.24 | valid ppl 1.27
-----------------------------------------------------------------------------------------
使用测试数据集评估模型¶
应用最佳模型以使用测试数据集检查结果。
test_loss = evaluate(best_model, test_data)
print('=' * 89)
print('| End of training | test loss {:5.2f} | test ppl {:8.2f}'.format(
test_loss, math.exp(test_loss)))
print('=' * 89)
=========================================================================================
| End of training | test loss 0.21 | test ppl 1.23
=========================================================================================
脚本总运行时间：（8 分钟 4.632 秒）
