微信扫码
添加专属顾问
 
                        我要投稿
探索RoPE如何让AI理解词语位置关系,这种创新的旋转编码技术正改变大模型处理序列的方式。 核心内容: 1. 传统位置编码方法的局限性分析 2. RoPE旋转位置编码的核心原理与数学表达 3. 实际代码实现与应用效果展示
 
                                在自注意力里,Q(查询向量)和 K(键向量)要做点积:
结果是什么?
相对位置关系自然保留:Q 和 K 的旋转差值就是它们的相对位置
外推能力强:没见过的长序列也能用,因为旋转是周期性的
# -*- coding: utf-8 -*-import torchimport torch.nn as nnimport torch.nn.functional as Fimport torch.optim as optimimport jiebaimport matplotlib.pyplot as pltimport numpy as npplt.rcParams['font.sans-serif'] = ['SimHei'] # 中文字体plt.rcParams['axes.unicode_minus'] = False# ===== 准备《水浒传》样本文本 =====text_samples = ["""张天师祈禳瘟疫,洪太尉误走妖魔。话说大宋天子仁宗皇帝在位年间,京师瘟疫流行,百姓多有染病。天子召张天师入宫祈禳,命洪太尉押送香火,不料误开封印,放出妖魔。""","""王教头私走延安府,九纹龙大闹史家村。史进自幼好武,学成十八般武艺,因打死恶霸,被官府缉拿。王进教头见势不妙,离开东京前往延安府,途经史家村。""","""史大郎夜走华阴县,鲁提辖拳打镇关西。史进与鲁达结义,路遇镇关西郑屠,见其欺压妇女,鲁达愤然出手,三拳打死郑屠,遂落草为寇。"""]# ===== 中文分词 =====def tokenize_texts(text_list):tokenized = []for t in text_list:words = list(jieba.cut(t))words = [w.strip() for w in words if w.strip()]tokenized.append(words)return tokenizedsentences = tokenize_texts(text_samples)# ===== 构建词表 =====vocab = {}for sent in sentences:for w in sent:if w not in vocab:vocab[w] = len(vocab)vocab["<PAD>"] = len(vocab)vocab_size = len(vocab)embed_dim = 32seq_len = max(len(s) for s in sentences)# 将句子转为索引,并paddef encode_sentences(sentences, vocab, seq_len):data = []for s in sentences:idxs = [vocab[w] for w in s]if len(idxs) < seq_len:idxs += [vocab["<PAD>"]] * (seq_len - len(idxs))data.append(idxs)return torch.tensor(data)input_ids = encode_sentences(sentences, vocab, seq_len)# ===== RoPE实现 =====def apply_rope(x):"""支持输入维度:- (B, T, D) 或- (B, T, H, D)返回相同形状,且对最后一维做 RoPE(要求 D 为偶数)"""orig_shape = x.shapeif len(orig_shape) == 3:# (B, T, D) -> 转为 (B, T, 1, D) 方便统一处理x = x.unsqueeze(2)squeezed = Trueelse:squeezed = False# 形状为 (B, T, H, D)# 现在 x.shape = (B, T, H, D)bsz, seqlen, nheads, head_dim = x.shapeassert head_dim % 2 == 0, "head_dim must be even for RoPE"device = x.devicedtype = x.dtypehalf = head_dim // 2# theta: (half,)theta = 10000 ** (-torch.arange(0, half, device=device, dtype=dtype) / half) # (half,)# seq positions: (seqlen,)seq_idx = torch.arange(seqlen, device=device, dtype=dtype) # (seqlen,)# freqs: (seqlen, half)freqs = torch.einsum('n,d->nd', seq_idx, theta)cos = freqs.cos().view(1, seqlen, 1, half) # (1, T, 1, half)sin = freqs.sin().view(1, seqlen, 1, half) # (1, T, 1, half)x1 = x[..., :half] # (B, T, H, half)x2 = x[..., half:] # (B, T, H, half)x_rotated = torch.cat([x1 * cos - x2 * sin,x1 * sin + x2 * cos], dim=-1) # (B, T, H, D)if squeezed:x_rotated = x_rotated.squeeze(2) # back to (B, T, D)return x_rotated# ===== 多头注意力 with RoPE =====class MultiHeadSelfAttentionRoPE(nn.Module):def __init__(self, embed_dim, num_heads, dropout=0.1):super().__init__()self.embed_dim = embed_dimself.num_heads = num_headsself.head_dim = embed_dim // num_headsself.dropout = dropoutself.q_proj = nn.Linear(embed_dim, embed_dim)self.k_proj = nn.Linear(embed_dim, embed_dim)self.v_proj = nn.Linear(embed_dim, embed_dim)self.out_proj = nn.Linear(embed_dim, embed_dim)self.last_attn_weights = Nonedef forward(self, x):B, T, C = x.size()q = self.q_proj(x).view(B, T, self.num_heads, self.head_dim)k = self.k_proj(x).view(B, T, self.num_heads, self.head_dim)v = self.v_proj(x).view(B, T, self.num_heads, self.head_dim)# 应用 RoPEq = apply_rope(q)k = apply_rope(k)# 注意力计算attn_scores = torch.einsum('bthd,bshd->bhts', q, k) / (self.head_dim ** 0.5)attn_weights = F.softmax(attn_scores, dim=-1)attn_weights = F.dropout(attn_weights, p=self.dropout, training=self.training)self.last_attn_weights = attn_weights.detach()out = torch.einsum('bhts,bshd->bthd', attn_weights, v)out = out.reshape(B, T, C)return self.out_proj(out)# ===== 模型训练 =====embedding = nn.Embedding(vocab_size, embed_dim)model = MultiHeadSelfAttentionRoPE(embed_dim, num_heads=4, dropout=0.1)criterion = nn.MSELoss()optimizer = optim.Adam(list(model.parameters()) + list(embedding.parameters()), lr=1e-3)epochs = 200for epoch in range(epochs):model.train()x = embedding(input_ids)target = x.clone()out = model(x)loss = criterion(out, target)optimizer.zero_grad()loss.backward()optimizer.step()if (epoch + 1) % 50 == 0:print(f"Epoch {epoch+1}, Loss: {loss.item():.6f}")# ===== 注意力热图可视化 =====def plot_attention(attn, sentence_tokens, filename):heads = attn.shape[0]fig, axes = plt.subplots(1, heads, figsize=(4*heads, 4))if heads == 1:axes = [axes]for h in range(heads):ax = axes[h]attn_head = attn[h].numpy()im = ax.imshow(attn_head, cmap='viridis')ax.set_xticks(np.arange(len(sentence_tokens)))ax.set_yticks(np.arange(len(sentence_tokens)))ax.set_xticklabels(sentence_tokens, rotation=90)ax.set_yticklabels(sentence_tokens)ax.set_title(f"Head {h+1}")fig.colorbar(im, ax=ax)plt.tight_layout()plt.savefig(filename)plt.close()model.eval()with torch.no_grad():x = embedding(input_ids)_ = model(x)attn_weights = model.last_attn_weights # (batch, heads, seq, seq)for i, tokens in enumerate(sentences):attn = attn_weights[i]plot_attention(attn.cpu(), tokens, f"rope_attention_sentence{i+1}.png")print("RoPE多头注意力热图已生成,文件名为 rope_attention_sentenceX.png")
53AI,企业落地大模型首选服务商
产品:场景落地咨询+大模型应用平台+行业解决方案
承诺:免费POC验证,效果达标后再合作。零风险落地应用大模型,已交付160+中大型企业
2025-10-31
Opera One升级内置AI 迎来智能助手新纪元
2025-10-31
LangExtract——大模型文本提炼工具
2025-10-31
用户测评|DeepSeek-OCR,你用了吗?
2025-10-31
从Palantir智能化技术路线看AI时代企业级架构平台的核心战略位置
2025-10-31
OpenAI 公开 Atlas 架构:为 Agent 重新发明浏览器
2025-10-31
Palantir 本体论模式:重塑企业 AI 应用的 “语义根基” 与产业启示
2025-10-31
树莓派这种“玩具级”设备,真能跑大模型吗?
2025-10-30
Cursor 2.0的一些有趣的新特性
 
            2025-08-21
2025-08-21
2025-08-19
2025-09-16
2025-10-02
2025-09-08
2025-09-17
2025-08-19
2025-09-29
2025-08-20