Personagpt

🚀 多人格对话代理（PersonaGPT）

PersonaGPT 是一个开放领域的对话代理，旨在完成两项任务：

1. 根据输入的个性事实（机器人的“人设”配置文件）解码出“个性化”回复。
2. 通过“动作代码”（例如，“谈论工作”“询问喜欢的音乐”）将“轮次级目标”融入其回复中。

它基于 [GPT - 2](https://github.com/openai/gpt - 2) 架构的 [DialoGPT - medium](https://huggingface.co/microsoft/DialoGPT - medium) 预训练模型构建。该模型在 Persona - Chat 数据集上进行训练，并添加了特殊标记，以便在双人对话中更好地区分对话历史和个性特征。此外，还使用了一些主动学习方法来训练模型，使其能够根据轮次级目标进行“可控”解码。

📚 完整仓库

预处理、训练和实现细节可在 personaGPT 仓库 中找到。

🚀 快速开始

📦 安装

本项目依赖于 transformers 和 torch 库，你可以使用以下命令进行安装：

pip install transformers torch

💻 使用示例

基础用法

1. 加载模型并定义一些辅助函数：

from transformers import GPT2Tokenizer, GPT2LMHeadModel
import torch
tokenizer = GPT2Tokenizer.from_pretrained("af1tang/personaGPT")
model = GPT2LMHeadModel.from_pretrained("af1tang/personaGPT")
if torch.cuda.is_available():
    model = model.cuda()
## utility functions ##
flatten = lambda l: [item for sublist in l for item in sublist]

def to_data(x):
    if torch.cuda.is_available():
        x = x.cpu()
    return x.data.numpy()

def to_var(x):
    if not torch.is_tensor(x):
        x = torch.Tensor(x)
    if torch.cuda.is_available():
        x = x.cuda()
    return x

def display_dialog_history(dialog_hx):
    for j, line in enumerate(dialog_hx):
        msg = tokenizer.decode(line)
        if j %2 == 0:
            print(">> User: "+ msg)
        else:
            print("Bot: "+msg)
            print()

def generate_next(bot_input_ids, do_sample=True, top_k=10, top_p=.92,
                  max_length=1000, pad_token=tokenizer.eos_token_id):
    full_msg = model.generate(bot_input_ids, do_sample=True,
                                              top_k=top_k, top_p=top_p, 
                                              max_length=max_length, pad_token_id=tokenizer.eos_token_id)
    msg = to_data(full_msg.detach()[0])[bot_input_ids.shape[-1]:]
    return msg

2. 为你的聊天机器人伙伴设置一组个性特征：

# get personality facts for conversation
personas = []
for i in range(3):
    response = input(">> Fact %d: "%(i+1))+ tokenizer.eos_token
    personas.append(response)
personas = tokenizer.encode(''.join(['<|p2|>'] + personas + ['<|sep|>'] + ['<|start|>']))

3. 个性化对话生成：使用以下循环与模型进行交互：

# converse for 8 turns
dialog_hx = []
for step in range(8):
    # encode the user input
    user_inp = tokenizer.encode(input(">> User: ") + tokenizer.eos_token)
    # append to the chat history
    dialog_hx.append(user_inp)
        
    # generated a response while limiting the total chat history to 1000 tokens, 
    bot_input_ids = to_var([personas + flatten(dialog_hx)]).long()
    msg = generate_next(bot_input_ids)
    dialog_hx.append(msg)
    print("Bot: {}".format(tokenizer.decode(msg, skip_special_tokens=True)))

以下是个性化解码的示例：

4. 可控回复生成：使用以下程序与模型进行交互：

## available actions ##
action_space = [ 'ask about kids.', "ask about pets.", 'talk about work.', 
               'ask about marital status.', 'talk about travel.', 'ask about age and gender.',
        'ask about hobbies.', 'ask about favorite food.', 'talk about movies.', 
        'talk about music.', 'talk about politics.']
# converse for 8 turns
dialog_hx = []
for step in range(8):
    # choose an action
    act = None
    while act not in action_space:
        display_dialog_history(dialog_hx)
        print()
        print(" actions: ")
        for k,v in enumerate(action_space): print(k,v)
        try:
            act = action_space[int(input(" input [0-10]: " ))]
        except:
            act = None
    print()
    # format into prefix code
    action_prefix = tokenizer.encode(''.join(['<|act|> '] + [act] + ['<|p1|>'] + [] + ['<|sep|>'] + ['<|start|>']))
    bot_input_ids = to_var([action_prefix + flatten(dialog_hx)]).long()
    
    # generate query conditioned on action
    msg = generate_next(bot_input_ids, top_k=top_k, top_p=top_p, max_length=max_length)
    dialog_hx.append(msg)
    
    # generate bot response
    bot_input_ids = to_var([personas+ flatten(dialog_hx)]).long()
    msg = generate_next(bot_input_ids, top_k=top_k, top_p=top_p, max_length=max_length)
    dialog_hx.append(msg)
display_dialog_history(dialog_hx)

以下是可控回复生成的示例：

📄 许可证

本项目采用 GPL - 3.0 许可证。