缩略图: https://cdn.discordapp.com/app-icons/870239976690970625/c02cae78ae105f07969cfd8f8ea3d0a0.png 标签:

• 对话式 许可证: mit

此AI模型已过时。请查看Aeona

基于microsoft/DialoGPT-small开发的生成式AI。

训练数据来源：

 https://www.kaggle.com/Cornell-University/movie-dialog-corpus

 https://www.kaggle.com/jef1056/discord-data

在线演示

示例代码：

from transformers import AutoTokenizer, AutoModelWithLMHead
  
tokenizer = AutoTokenizer.from_pretrained("deepparag/DumBot")
model = AutoModelWithLMHead.from_pretrained("deepparag/DumBot")
# 进行4轮对话
for step in range(4):
    # 编码用户输入，添加eos_token并返回PyTorch张量
    new_user_input_ids = tokenizer.encode(input(">> 用户：") + tokenizer.eos_token, return_tensors='pt')
    # print(new_user_input_ids)
    # 将用户输入添加到对话历史中
    bot_input_ids = torch.cat([chat_history_ids, new_user_input_ids], dim=-1) if step > 0 else new_user_input_ids
    # 生成响应，限制总对话历史不超过1000个token
    chat_history_ids = model.generate(
        bot_input_ids, max_length=200,
        pad_token_id=tokenizer.eos_token_id,  
        no_repeat_ngram_size=4,       
        do_sample=True, 
        top_k=100, 
        top_p=0.7,
        temperature=0.8
    )
    
    # 优雅地打印机器人最后输出的token
    print("DumBot：{}".format(tokenizer.decode(chat_history_ids[:, bot_input_ids.shape[-1]:][0], skip_special_tokens=True)))