A new microblog sentiment recognition model has been added, based on the fine-tuning of GPT2, but it has not yet been adapted to the system.

model3/README.md

@@ -0,0 +1,93 @@
+# 微博情感识别模型-GPT2-Adapter微调
+
+## 项目说明
+这是一个基于GPT2的微博情感二分类模型，采用Adapter微调技术。通过Adapter微调，只需训练少量参数就可以让模型适应情感分析任务，大幅降低计算资源需求和模型体积。
+
+## 数据集
+使用微博情感数据集(weibo_senti_100k)，包含约10万条带情感标注的微博内容，正负向评论各约5万条。数据集标签：
+- 标签0：负面情感
+- 标签1：正面情感
+
+## 文件结构
+```
+GPT2-Adpter-tuning/
+├── adapter.py              # Adapter层的实现
+├── gpt2_adapter.py         # 针对GPT2模型的Adapter实现
+├── train.py                # 训练脚本
+├── predict.py              # 简化版预测脚本（交互式使用）
+├── models/                 # 本地存储的预训练模型
+│   └── gpt2-chinese/       # 中文GPT2模型及配置
+├── dataset/                # 数据集目录
+│   └── weibo_senti_100k.csv  # 微博情感数据集
+└── best_weibo_sentiment_model.pth  # 训练好的最佳模型
+```
+
+## 技术特点
+
+1. **参数高效微调**：相比全参数微调，仅训练约3%的参数
+2. **模型性能保持**：在仅训练少量参数的情况下，保持良好的分类性能
+3. **适用于资源受限环境**：模型体积小，推理速度快
+
+## 环境依赖
+- Python 3.6+
+- PyTorch
+- Transformers
+- Pandas
+- NumPy
+- Scikit-learn
+- Tqdm
+
+## 使用方法
+
+### 训练模型
+```bash
+python train.py
+```
+训练过程会自动：
+- 下载并本地保存中文GPT2预训练模型
+- 加载微博情感数据集
+- 训练模型并保存最佳模型
+
+### 情感分析预测
+```bash
+python predict.py
+```
+运行后将进入交互模式：
+- 在控制台输入要分析的微博文本
+- 系统会返回情感分析结果（正面/负面）和置信度
+- 输入'q'退出程序
+
+## 模型结构
+- 基础模型：`uer/gpt2-chinese-cluecorpussmall`中文预训练模型
+- 模型本地保存路径：`./models/gpt2-chinese/`
+- 通过在每个GPT2Block后添加Adapter层进行微调
+- 冻结原始GPT2参数，仅训练分类器和Adapter层参数
+
+## Adapter技术
+Adapter是一种参数高效的微调技术，通过在Transformer层中插入小型的瓶颈层，实现用少量参数适应下游任务的目的。主要特点：
+
+1. **参数高效**：相比全参数微调，Adapter只需训练很小一部分参数
+2. **防止遗忘**：保持原始预训练模型的参数不变，避免灾难性遗忘
+3. **适应多任务**：可以为不同任务训练不同的Adapter，共享同一个基础模型
+
+在本项目中，我们在每个GPT2Block后添加了一个Adapter层，Adapter的隐藏层大小为64，远小于原始模型的隐藏层大小（通常为768或1024）。
+
+## 使用示例
+```
+使用设备: cuda
+加载模型: best_weibo_sentiment_model.pth
+
+============= 微博情感分析 =============
+输入微博内容进行分析 (输入 'q' 退出):
+
+请输入微博内容: 这部电影真是太好看了，我非常喜欢！
+预测结果: 正面情感 (置信度: 0.9876)
+
+请输入微博内容: 服务态度差，价格还贵，一点都不推荐
+预测结果: 负面情感 (置信度: 0.9742)
+```
+
+## 注意事项
+- 预测脚本使用本地模型路径，不需要在线下载模型
+- 确保`models/gpt2-chinese/`目录包含从训练过程中保存的模型文件
+- 首次运行train.py时会自动下载并保存模型，请确保网络连接 

model3/adapter.py

@@ -0,0 +1,42 @@
+import torch
+import torch.nn as nn
+
+class AdapterLayer(nn.Module):
+    """
+    Adapter层实现
+    将其添加到Transformer层中可以实现参数高效微调
+    """
+    def __init__(self, input_size, adapter_size):
+        super(AdapterLayer, self).__init__()
+        # 降维全连接层
+        self.down_project = nn.Linear(input_size, adapter_size)
+        # 激活函数
+        self.activation = nn.ReLU()
+        # 升维全连接层
+        self.up_project = nn.Linear(adapter_size, input_size)
+        
+        # 初始化参数
+        self._init_weights()
+    
+    def _init_weights(self):
+        # 初始化down_project用较小的值
+        nn.init.normal_(self.down_project.weight, std=1e-2)
+        nn.init.zeros_(self.down_project.bias)
+        
+        # 初始化up_project为接近零的值，确保训练初期对原始模型影响较小
+        nn.init.normal_(self.up_project.weight, std=1e-2)
+        nn.init.zeros_(self.up_project.bias)
+    
+    def forward(self, x):
+        # 保存原始输入用于残差连接
+        residual = x
+        
+        # 通过降维层
+        x = self.down_project(x)
+        # 激活
+        x = self.activation(x)
+        # 通过升维层
+        x = self.up_project(x)
+        
+        # 残差连接
+        return residual + x 

model3/dataset/intro.ipynb

@@ -0,0 +1,280 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# weibo_senti_100k 说明\n",
+    "0. **下载地址：** [百度网盘](https://pan.baidu.com/s/1DoQbki3YwqkuwQUOj64R_g)\n",
+    "1. **数据概览：** 10 万多条，带情感标注 新浪微博，正负向评论约各 5 万条\n",
+    "2. **推荐实验：** 情感/观点/评论 倾向性分析\n",
+    "2. **数据来源：** [新浪微博](https://weibo.com/)\n",
+    "3. **原数据集：** [新浪微博，情感分析标记语料共12万条](https://download.csdn.net/download/weixin_38442818/10214750)，网上搜集，具体作者、来源不详\n",
+    "4. **加工处理：**\n",
+    "    1. 将原来的 2 份文档，整合成 1 份 csv 文件\n",
+    "    2. 编码统一为 UTF-8\n",
+    "    3. 去重"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "path = 'weibo_senti_100k_文件夹_所在_路径'"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# 1. weibo_senti_100k.csv"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 加载数据"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "评论数目（总体）：119988\n",
+      "评论数目（正向）：59993\n",
+      "评论数目（负向）：59995\n"
+     ]
+    }
+   ],
+   "source": [
+    "pd_all = pd.read_csv(path + 'weibo_senti_100k.csv')\n",
+    "\n",
+    "print('评论数目（总体）：%d' % pd_all.shape[0])\n",
+    "print('评论数目（正向）：%d' % pd_all[pd_all.label==1].shape[0])\n",
+    "print('评论数目（负向）：%d' % pd_all[pd_all.label==0].shape[0])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 字段说明\n",
+    "\n",
+    "| 字段 | 说明 |\n",
+    "| ---- | ---- |\n",
+    "| label | 1 表示正向评论，0 表示负向评论 |\n",
+    "| review | 微博内容 |"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>label</th>\n",
+       "      <th>review</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>62050</th>\n",
+       "      <td>0</td>\n",
+       "      <td>太过分了@Rexzhenghao  //@Janie_Zhang:招行最近负面新闻越来越多呀...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>68263</th>\n",
+       "      <td>0</td>\n",
+       "      <td>希望你?得好?我本＂?肥血?史＂[晕][哈哈]@Pete三姑父</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>81472</th>\n",
+       "      <td>0</td>\n",
+       "      <td>有点想参加????[偷?]想安排下时间再决定[抓狂]//@黑晶晶crystal: @细腿大羽...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>42021</th>\n",
+       "      <td>1</td>\n",
+       "      <td>[给力]感谢所有支持雯婕的芝麻！[爱你]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7777</th>\n",
+       "      <td>1</td>\n",
+       "      <td>2013最后一天，在新加坡开心度过，向所有的朋友们问声：新年快乐！2014年，我们会更好[调...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>100399</th>\n",
+       "      <td>0</td>\n",
+       "      <td>大中午出门办事找错路，曝晒中。要多杯具有多杯具。[泪][泪][汗]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>82398</th>\n",
+       "      <td>0</td>\n",
+       "      <td>马航还会否认吗？到底在隐瞒啥呢？[抓狂]//@头条新闻: 转发微博</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>106423</th>\n",
+       "      <td>0</td>\n",
+       "      <td>克罗地亚球迷很爱放烟火！球又没进，就硝烟四起。[晕]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>24798</th>\n",
+       "      <td>1</td>\n",
+       "      <td>[抱抱]福芦 TangRoulou 吉祥书 8.8折优惠 &gt;&gt;&gt; http://t.cn/z...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6598</th>\n",
+       "      <td>1</td>\n",
+       "      <td>回复@钱旭明QXM:[嘻嘻][嘻嘻] //@钱旭明QXM:杨大哥[good][good][g...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>53920</th>\n",
+       "      <td>1</td>\n",
+       "      <td>人家这脸长的!!!!!![哈哈]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15587</th>\n",
+       "      <td>1</td>\n",
+       "      <td>这个价不算高，和一天内训相比相差无几。。[哈哈]//@博通传媒v: 6个月！一个月工资1万，...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>101237</th>\n",
+       "      <td>0</td>\n",
+       "      <td>终于收工啦，脚丫子快冻掉了[泪][泪][泪]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>82449</th>\n",
+       "      <td>0</td>\n",
+       "      <td>我决定从今天开始我想吃什么就去吃什么，一个人吃也无所谓，重点是不要因为别人的意见委屈了自己[...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>32537</th>\n",
+       "      <td>1</td>\n",
+       "      <td>飘雪的北京 需要双份早餐.......//@美食天下: [哈哈]//@王淼Margay: 屁...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10630</th>\n",
+       "      <td>1</td>\n",
+       "      <td>[耶]，这个太赞了，生活大爆炸第六季马上要出啦[鼓掌] //@-郑瑜-:这个不错 //@经典...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>85130</th>\n",
+       "      <td>0</td>\n",
+       "      <td>刚追完#倾世皇妃#，#千山暮雪#又紧随其后，网速和更新速度都太不给力，尽管我看过原著，还是焦...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>105956</th>\n",
+       "      <td>0</td>\n",
+       "      <td>晚上看金二胖?察前?，推出的火炮基座?糟了，可以PK了[泪] //@艾米粒er: //@wi...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>72391</th>\n",
+       "      <td>0</td>\n",
+       "      <td>必须把中国足球的伟大，用我的职业演说出来 //@袁腾飞:[泪]</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10761</th>\n",
+       "      <td>1</td>\n",
+       "      <td>[鼓掌] //@宁波香格里拉大酒店: 小编来答疑，周五晚惊艳全场的树根蛋糕到底有多长？蛋糕全...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "        label                                             review\n",
+       "62050       0  太过分了@Rexzhenghao  //@Janie_Zhang:招行最近负面新闻越来越多呀...\n",
+       "68263       0                    希望你?得好?我本＂?肥血?史＂[晕][哈哈]@Pete三姑父\n",
+       "81472       0  有点想参加????[偷?]想安排下时间再决定[抓狂]//@黑晶晶crystal: @细腿大羽...\n",
+       "42021       1                               [给力]感谢所有支持雯婕的芝麻！[爱你]\n",
+       "7777        1  2013最后一天，在新加坡开心度过，向所有的朋友们问声：新年快乐！2014年，我们会更好[调...\n",
+       "100399      0                  大中午出门办事找错路，曝晒中。要多杯具有多杯具。[泪][泪][汗]\n",
+       "82398       0                  马航还会否认吗？到底在隐瞒啥呢？[抓狂]//@头条新闻: 转发微博\n",
+       "106423      0                         克罗地亚球迷很爱放烟火！球又没进，就硝烟四起。[晕]\n",
+       "24798       1  [抱抱]福芦 TangRoulou 吉祥书 8.8折优惠 >>> http://t.cn/z...\n",
+       "6598        1  回复@钱旭明QXM:[嘻嘻][嘻嘻] //@钱旭明QXM:杨大哥[good][good][g...\n",
+       "53920       1                                   人家这脸长的!!!!!![哈哈]\n",
+       "15587       1  这个价不算高，和一天内训相比相差无几。。[哈哈]//@博通传媒v: 6个月！一个月工资1万，...\n",
+       "101237      0                             终于收工啦，脚丫子快冻掉了[泪][泪][泪]\n",
+       "82449       0  我决定从今天开始我想吃什么就去吃什么，一个人吃也无所谓，重点是不要因为别人的意见委屈了自己[...\n",
+       "32537       1  飘雪的北京 需要双份早餐.......//@美食天下: [哈哈]//@王淼Margay: 屁...\n",
+       "10630       1  [耶]，这个太赞了，生活大爆炸第六季马上要出啦[鼓掌] //@-郑瑜-:这个不错 //@经典...\n",
+       "85130       0  刚追完#倾世皇妃#，#千山暮雪#又紧随其后，网速和更新速度都太不给力，尽管我看过原著，还是焦...\n",
+       "105956      0  晚上看金二胖?察前?，推出的火炮基座?糟了，可以PK了[泪] //@艾米粒er: //@wi...\n",
+       "72391       0                    必须把中国足球的伟大，用我的职业演说出来 //@袁腾飞:[泪]\n",
+       "10761       1  [鼓掌] //@宁波香格里拉大酒店: 小编来答疑，周五晚惊艳全场的树根蛋糕到底有多长？蛋糕全..."
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pd_all.sample(20)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.4"
+  },
+  "widgets": {
+   "state": {},
+   "version": "1.1.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

model3/example/Adapter.py

@@ -0,0 +1,20 @@
+import torch
+import torch.nn as nn
+
+class AdapterLayer(nn.Module):
+    def __init__(self, input_size, adapter_size):
+        super(AdapterLayer, self).__init__()
+        # 第一个全连接层降维
+        self.down_project = nn.Linear(input_size, adapter_size)
+        # ReLU激活函数
+        self.relu = nn.ReLU()
+        # 第二个全连接层升维
+        self.up_project = nn.Linear(adapter_size, input_size)
+
+    def forward(self, x):
+        # 通过Adapter层的前向传播
+        down_projected = self.down_project(x)
+        relu = self.relu(down_projected)
+        up_projected = self.up_project(x)
+        # 将Adapter的输出与输入相加（残差连接）
+        return x + up_projected

model3/example/README.md

@@ -0,0 +1,11 @@
+一种Adapter-tuning的实现方式，只提供的思路，具体可以视情况稍微修改。
+
+
+这里补充一些模型层数：
+GPT-2 Small：12个GPT2Block，约有1.17亿个参数。
+GPT-2 Medium：24个GPT2Block，约有3.48亿个参数。
+GPT-2 Large：36个GPT2Block，约有7.55亿个参数。
+GPT-2 XL (也称为Extra Large)：48个GPT2Block，约有15.54亿个参数。
+
+RoBERTa Base：12个RobertaLayer，总共约有1.25亿个参数。
+RoBERTa Large：24个RobertaLayer，总共约有3.55亿个参数。

model3/example/Roberta_adapter.py

@@ -0,0 +1,22 @@
+from transformers.models.roberta.modeling_roberta import RobertaLayer
+
+class RobertaLayerWithAdapter(RobertaLayer):
+    def __init__(self, config):
+        super().__init__(config)
+        # 假设Adapter的大小为64
+        adapter_size = 64
+        self.adapter = AdapterLayer(config.hidden_size, adapter_size)
+
+    def forward(self, hidden_states, attention_mask=None, head_mask=None, encoder_hidden_states=None, encoder_attention_mask=None, past_key_value=None, output_attentions=False):
+        # 调用原始的前向传播方法
+        self_outputs = super().forward(hidden_states, attention_mask, head_mask, encoder_hidden_states, encoder_attention_mask, past_key_value, output_attentions)
+        # 得到Transformer层的输出
+        sequence_output = self_outputs[0]
+        # 将输出通过Adapter层
+        sequence_output = self.adapter(sequence_output)
+        # 返回修改后的输出（其他输出保持不变）
+        return (sequence_output,) + self_outputs[1:]
+
+"""
+RoBERTa的每个RobertaLayer包含一个自注意力（self-attention）机制和一个前馈网络，这些层共同构成了RoBERTa的基础架构。
+"""

model3/example/gpt2_adapter.py

@@ -0,0 +1,40 @@
+from transformers.models.gpt2.modeling_gpt2 import GPT2Block
+
+class GPT2BlockWithAdapter(GPT2Block):
+    def __init__(self, config):
+        super().__init__(config)
+        # 假设Adapter的大小为64
+        adapter_size = 64
+        self.adapter = AdapterLayer(config.n_embd, adapter_size)
+
+    def forward(
+        self,
+        hidden_states,
+        layer_past=None,
+        attention_mask=None,
+        head_mask=None,
+        use_cache=False,
+        output_attentions=False,
+    ):
+        # 调用原始的前向传播方法
+        attn_outputs = super().forward(
+            hidden_states,
+            layer_past=layer_past,
+            attention_mask=attention_mask,
+            head_mask=head_mask,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+        )
+        # 得到Transformer层的输出
+        a = attn_outputs[0]  # 输出的第一部分是attention的结果
+        # 将输出通过Adapter层
+        a = self.adapter(a)
+        # 返回修改后的输出（其他输出保持不变）
+        outputs = (a,) + attn_outputs[1:]
+        return outputs
+"""
+每个GPT2Block包含了一系列的自注意力（Self-Attention）和前馈网络（Feed-Forward）层，这些层共同构成了模型的基础架构。
+
+"""
+
+

model3/gpt2_adapter.py

@@ -0,0 +1,60 @@
+import torch
+import torch.nn as nn
+from transformers.models.gpt2.modeling_gpt2 import GPT2Block
+from adapter import AdapterLayer
+
+class GPT2BlockWithAdapter(nn.Module):
+    """
+    带Adapter的GPT2Block层
+    在原始GPT2Block的基础上添加Adapter层实现参数高效微调
+    """
+    def __init__(self, config):
+        super(GPT2BlockWithAdapter, self).__init__()
+        # 创建标准的GPT2Block
+        self.original_block = GPT2Block(config)
+        
+        # 添加Adapter层
+        adapter_size = 64  # Adapter的隐藏层大小
+        self.adapter = AdapterLayer(config.hidden_size, adapter_size)
+    
+    def forward(
+        self,
+        hidden_states,
+        layer_past=None,
+        attention_mask=None,
+        head_mask=None,
+        encoder_hidden_states=None,
+        encoder_attention_mask=None,
+        use_cache=False,
+        output_attentions=False,
+        **kwargs  # 使用**kwargs接收所有其他参数
+    ):
+        # 首先通过原始的GPT2Block，只传递它支持的参数
+        outputs = self.original_block(
+            hidden_states,
+            layer_past=layer_past,
+            attention_mask=attention_mask,
+            head_mask=head_mask,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_attention_mask,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+        )
+        
+        # 原始输出中的第一个元素是隐藏状态
+        hidden_states = outputs[0]
+        
+        # 将隐藏状态通过Adapter层
+        hidden_states = self.adapter(hidden_states)
+        
+        # 更新输出的隐藏状态
+        outputs = (hidden_states,) + outputs[1:]
+        
+        return outputs
+    
+    def load_state_dict(self, state_dict, strict=True):
+        """
+        自定义加载参数方法，用于从原始GPT2Block加载参数
+        """
+        # 将所有参数传递给原始Block
+        return self.original_block.load_state_dict(state_dict, strict=strict) 

model3/predict.py

@@ -0,0 +1,61 @@
+import torch
+from transformers import BertTokenizer
+from train import GPT2ClassifierWithAdapter
+
+def main():
+    # 设置设备
+    device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+    print(f"使用设备: {device}")
+    
+    # 使用本地模型路径而不是在线模型名称
+    local_model_path = './models/gpt2-chinese'
+    model_path = 'best_weibo_sentiment_model.pth'
+    
+    print(f"加载模型: {model_path}")
+    # 从本地加载tokenizer
+    tokenizer = BertTokenizer.from_pretrained(local_model_path)
+    if tokenizer.pad_token is None:
+        tokenizer.pad_token = '[PAD]'
+    
+    # 加载模型，使用本地模型路径
+    model = GPT2ClassifierWithAdapter(local_model_path)
+    model.load_state_dict(torch.load(model_path, map_location=device))
+    model.to(device)
+    model.eval()
+    
+    print("\n============= 微博情感分析 =============")
+    print("输入微博内容进行分析 (输入 'q' 退出):")
+    
+    while True:
+        text = input("\n请输入微博内容: ")
+        if text.lower() == 'q':
+            break
+        
+        # 对文本进行编码
+        encoding = tokenizer(
+            text,
+            max_length=128,
+            padding='max_length',
+            truncation=True,
+            return_tensors='pt'
+        )
+        
+        # 转移到设备
+        input_ids = encoding['input_ids'].to(device)
+        attention_mask = encoding['attention_mask'].to(device)
+        
+        # 预测
+        with torch.no_grad():
+            outputs = model(input_ids=input_ids, attention_mask=attention_mask)
+            logits = outputs.logits
+            probabilities = torch.softmax(logits, dim=1)
+            prediction = torch.argmax(probabilities, dim=1).item()
+        
+        # 输出结果
+        confidence = probabilities[0][prediction].item()
+        label = "正面情感" if prediction == 1 else "负面情感"
+        
+        print(f"预测结果: {label} (置信度: {confidence:.4f})")
+
+if __name__ == "__main__":
+    main() 

model3/train.py

@@ -0,0 +1,310 @@
+import os
+import random
+import numpy as np
+import pandas as pd
+import torch
+import torch.nn as nn
+from torch.utils.data import Dataset, DataLoader
+from transformers import GPT2Config, GPT2ForSequenceClassification, BertTokenizer, get_linear_schedule_with_warmup
+from torch.optim import AdamW
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import accuracy_score, f1_score
+from tqdm import tqdm
+
+from adapter import AdapterLayer
+from gpt2_adapter import GPT2BlockWithAdapter
+
+# 设置随机种子
+def set_seed(seed):
+    random.seed(seed)
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    if torch.cuda.is_available():
+        torch.cuda.manual_seed_all(seed)
+
+set_seed(42)
+
+# 定义微博情感分析数据集
+class WeiboSentimentDataset(Dataset):
+    def __init__(self, reviews, labels, tokenizer, max_length=128):
+        self.reviews = reviews
+        self.labels = labels
+        self.tokenizer = tokenizer
+        self.max_length = max_length
+        
+    def __len__(self):
+        return len(self.reviews)
+    
+    def __getitem__(self, idx):
+        review = str(self.reviews[idx])
+        label = self.labels[idx]
+        
+        encoding = self.tokenizer(
+            review,
+            max_length=self.max_length,
+            padding='max_length',
+            truncation=True,
+            return_tensors='pt'
+        )
+        
+        return {
+            'input_ids': encoding['input_ids'].flatten(),
+            'attention_mask': encoding['attention_mask'].flatten(),
+            'labels': torch.tensor(label, dtype=torch.long)
+        }
+
+# 定义GPT2分类模型，带Adapter
+class GPT2ClassifierWithAdapter(nn.Module):
+    def __init__(self, pretrained_model_name, num_labels=2):
+        super(GPT2ClassifierWithAdapter, self).__init__()
+        # 加载预训练模型
+        self.gpt2 = GPT2ForSequenceClassification.from_pretrained(
+            pretrained_model_name,
+            num_labels=num_labels
+        )
+        
+        # 确保模型配置中设置了pad_token_id
+        self.gpt2.config.pad_token_id = self.gpt2.config.eos_token_id
+        
+        # 替换原始的GPT2Block为带Adapter的版本
+        config = self.gpt2.config
+        for i in range(len(self.gpt2.transformer.h)):
+            # 保存原始权重
+            old_block = self.gpt2.transformer.h[i]
+            # 创建带Adapter的新Block
+            new_block = GPT2BlockWithAdapter(config)
+            # 复制原始权重
+            new_block.load_state_dict(old_block.state_dict(), strict=False)
+            # 替换
+            self.gpt2.transformer.h[i] = new_block
+            
+        # 冻结原始GPT2参数
+        for param in self.gpt2.parameters():
+            param.requires_grad = False
+            
+        # 解冻分类器层和Adapter层参数
+        for param in self.gpt2.score.parameters():
+            param.requires_grad = True
+            
+        # 解冻所有Adapter层
+        for i in range(len(self.gpt2.transformer.h)):
+            for param in self.gpt2.transformer.h[i].adapter.parameters():
+                param.requires_grad = True
+    
+    def forward(self, input_ids, attention_mask, labels=None):
+        return self.gpt2(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            labels=labels
+        )
+
+# 训练函数
+def train_model(model, train_dataloader, val_dataloader, optimizer, scheduler, device, epochs=3):
+    best_f1 = 0.0
+    
+    for epoch in range(epochs):
+        print(f"======== Epoch {epoch+1} / {epochs} ========")
+        model.train()
+        total_loss = 0
+        
+        # 训练循环
+        progress_bar = tqdm(train_dataloader, desc="Training", position=0, leave=True)
+        for batch in progress_bar:
+            # 将数据移到GPU
+            batch = {k: v.to(device) for k, v in batch.items()}
+            
+            # 清零梯度
+            optimizer.zero_grad()
+            
+            # 前向传播
+            outputs = model(
+                input_ids=batch['input_ids'],
+                attention_mask=batch['attention_mask'],
+                labels=batch['labels']
+            )
+            
+            loss = outputs.loss
+            total_loss += loss.item()
+            
+            # 反向传播
+            loss.backward()
+            
+            # 梯度裁剪，防止梯度爆炸
+            torch.nn.utils.clip_grad_norm_(model.parameters(), 1.0)
+            
+            # 参数更新
+            optimizer.step()
+            scheduler.step()
+            
+            # 更新进度条
+            progress_bar.set_postfix({"loss": loss.item()})
+        
+        # 计算平均训练损失
+        avg_train_loss = total_loss / len(train_dataloader)
+        print(f"Average training loss: {avg_train_loss:.4f}")
+        
+        # 评估模型
+        val_metrics = evaluate_model(model, val_dataloader, device)
+        print(f"Validation Loss: {val_metrics['loss']:.4f}")
+        print(f"Validation Accuracy: {val_metrics['accuracy']:.4f}")
+        print(f"Validation F1 Score: {val_metrics['f1']:.4f}")
+        
+        # 保存最佳模型
+        if val_metrics['f1'] > best_f1:
+            best_f1 = val_metrics['f1']
+            torch.save(model.state_dict(), "best_weibo_sentiment_model.pth")
+            print("Saved best model!")
+
+# 评估函数
+def evaluate_model(model, dataloader, device):
+    model.eval()
+    total_loss = 0
+    all_preds = []
+    all_labels = []
+    
+    with torch.no_grad():
+        for batch in tqdm(dataloader, desc="Evaluating"):
+            batch = {k: v.to(device) for k, v in batch.items()}
+            
+            outputs = model(
+                input_ids=batch['input_ids'],
+                attention_mask=batch['attention_mask'],
+                labels=batch['labels']
+            )
+            
+            loss = outputs.loss
+            total_loss += loss.item()
+            
+            # 获取预测结果
+            logits = outputs.logits
+            preds = torch.argmax(logits, dim=1).cpu().numpy()
+            labels = batch['labels'].cpu().numpy()
+            
+            all_preds.extend(preds)
+            all_labels.extend(labels)
+    
+    # 计算评估指标
+    accuracy = accuracy_score(all_labels, all_preds)
+    f1 = f1_score(all_labels, all_preds, average='macro')
+    avg_loss = total_loss / len(dataloader)
+    
+    return {
+        'loss': avg_loss,
+        'accuracy': accuracy,
+        'f1': f1
+    }
+
+def main():
+    # 设置模型本地保存路径
+    model_name = 'uer/gpt2-chinese-cluecorpussmall'
+    local_model_path = './models/gpt2-chinese'
+    
+    # 确保目录存在
+    os.makedirs(local_model_path, exist_ok=True)
+    
+    # 加载数据集
+    print("加载微博情感数据集...")
+    df = pd.read_csv('dataset/weibo_senti_100k.csv')
+    
+    # 分割数据集
+    train_df, val_df = train_test_split(df, test_size=0.1, random_state=42, stratify=df['label'])
+    
+    # 加载tokenizer和模型
+    print("加载预训练模型和tokenizer...")
+    
+    # 检查本地是否已有模型
+    if os.path.exists(os.path.join(local_model_path, 'config.json')):
+        print(f"从本地路径加载模型: {local_model_path}")
+        tokenizer = BertTokenizer.from_pretrained(local_model_path)
+    else:
+        print(f"从Hugging Face下载模型到: {local_model_path}")
+        tokenizer = BertTokenizer.from_pretrained(model_name, cache_dir=local_model_path)
+        # 保存tokenizer到本地
+        tokenizer.save_pretrained(local_model_path)
+    
+    # 设置padding token (BertTokenizer通常已有[PAD]作为padding token)
+    if tokenizer.pad_token is None:
+        # 如果没有，显式设置为[PAD]
+        tokenizer.pad_token = '[PAD]'
+    
+    # 记录pad_token的ID，确保模型和tokenizer使用相同的pad_token_id
+    pad_token_id = tokenizer.pad_token_id
+    
+    # 创建数据集
+    train_dataset = WeiboSentimentDataset(
+        train_df['review'].values,
+        train_df['label'].values,
+        tokenizer
+    )
+    
+    val_dataset = WeiboSentimentDataset(
+        val_df['review'].values,
+        val_df['label'].values,
+        tokenizer
+    )
+    
+    # 创建数据加载器
+    train_dataloader = DataLoader(train_dataset, batch_size=16, shuffle=True)
+    val_dataloader = DataLoader(val_dataset, batch_size=16)
+    
+    # 设置设备
+    device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+    print(f"使用设备: {device}")
+    
+    # 初始化模型
+    if (os.path.exists(os.path.join(local_model_path, 'pytorch_model.bin')) or 
+        os.path.exists(os.path.join(local_model_path, 'model.safetensors'))):
+        print(f"从本地路径加载模型权重: {local_model_path}")
+        model = GPT2ClassifierWithAdapter(local_model_path)
+    else:
+        print(f"从Hugging Face下载模型权重到: {local_model_path}")
+        # 直接从Hugging Face下载并保存完整模型
+        temp_model = GPT2ForSequenceClassification.from_pretrained(model_name)
+        temp_model.save_pretrained(local_model_path)
+        # 然后用保存的模型创建GPT2ClassifierWithAdapter
+        model = GPT2ClassifierWithAdapter(local_model_path)
+    
+    # 确保模型使用与tokenizer相同的pad_token_id
+    model.gpt2.config.pad_token_id = pad_token_id
+    model.to(device)
+    
+    # 统计需要训练的参数
+    total_params = sum(p.numel() for p in model.parameters())
+    trainable_params = sum(p.numel() for p in model.parameters() if p.requires_grad)
+    
+    print(f"模型总参数量: {total_params}")
+    print(f"需要训练的参数量: {trainable_params} ({trainable_params/total_params*100:.2f}%)")
+    
+    # 设置优化器和学习率调度器
+    optimizer = AdamW(
+        [p for p in model.parameters() if p.requires_grad],
+        lr=5e-5,
+        eps=1e-8
+    )
+    
+    # 设置总训练步数和warmup步数
+    total_steps = len(train_dataloader) * 2  # 2个epoch
+    warmup_steps = int(total_steps * 0.1)  # 10%的warmup
+    
+    scheduler = get_linear_schedule_with_warmup(
+        optimizer,
+        num_warmup_steps=warmup_steps,
+        num_training_steps=total_steps
+    )
+    
+    # 训练模型
+    print("开始训练...")
+    train_model(
+        model=model,
+        train_dataloader=train_dataloader,
+        val_dataloader=val_dataloader,
+        optimizer=optimizer,
+        scheduler=scheduler,
+        device=device,
+        epochs=2
+    )
+    
+    print("训练完成!")
+
+if __name__ == "__main__":
+    main()