Comments (4)
@chenhehong 谢谢,请问https://www.modelscope.cn/studios/damo/role_play_chat/files 这里面的roles_json和源码 有公开吗?想自己跑起来试一下
app_config.py
import os
import json
import requests
class AppConfig(object):
ALREADY_CONVERTED_MARK = "<!-- ALREADY CONVERTED BY PARSER. -->"
with open(os.path.join(os.path.dirname(__file__), 'main.css'), "r", encoding="utf-8") as f:
MAIN_CSS_CODE = f.read()
roles_json = os.getenv('roles_json')
print('roles json', roles_json)
ROLES = json.loads(requests.get(roles_json).content.decode('utf-8'))
from chatplug.
你好,这个是代码支持cpu推理的bug,mass下一个sdk版本会修复该问题。
当前版本临时的解决方法是将 modelscope/pipelines/nlp/fid_dialogue_pipeline.py 文件194-195行中if torch.cuda.is_available(): hypotheses = inputs.sequences.detach().cpu().tolist()
修改为hypotheses = inputs.sequences.detach().cpu().tolist()
from chatplug.
# 支持输入多段外部知识文本,进行知识增强
know_list = [
"《狂飙》由徐纪周执导的。《狂飙》的导演徐纪周也是编剧之一,代表作品有《永不磨灭的番号》《特战荣耀》《心理罪之城市之光》《杀虎口》《胭脂》等",
"《狂飙》(The Knockout)是一部由 张译、张颂文、李一桐、张志坚 领衔主演,韩童生 特邀主演,吴健、郝平 友情出演,高叶、贾冰、李健 主演,徐纪周 执导,朱俊懿、徐纪周 担任总编剧的 刑侦",
"狂飙是由**政法委宣传教育局,**政法委政法信息中心指导,爱奇艺,留白影视出品,徐纪周执导,张译,李一桐,张志坚领衔主演的刑侦剧。不是。是徐纪周,1976年12月19日出生,毕业于**戏剧"
]
input = {
"history": "你好[SEP]你好!很高兴与你交流![SEP]狂飙的导演是谁呀",
"bot_profile": "我是达摩院的语言模型ChatPLUG, 是基于海量数据训练得到。",
# 推理时,会根据[SEP]符号拆分为多段文本,分别和context合并后输入模型中,以节省计算量
"knowledge": "[SEP]".join(know_list),
}
谢谢 @chenhehong 的回复,我还有几个剖一,不知道是不是可以帮忙回答,谢谢:
- 官方示例中的 know_list 的知识增强最大支持多少内容? 我试过2w个字 64G的内存受不了了
- 我如果想对本地知识库进行问答可以基于know_list做吗?
- 我看到官方readme里面好像还能支持搜索引擎插件, 这个是如何实现的?
from chatplug.
- 知识增强最大支持的token数和机器显存大小有关,我们的推理测试一般只用千级别的token数,2w个字64G应该吃不消
- 本地知识库进行问答可以基于know_list>>可以
- chatplug是调用夸克api接口实现的,该api接口是付费接口,暂时提供不了
from chatplug.
Related Issues (15)
- [Question]:如何让模型能以stream方式输出? HOT 3
- [Question]:可以公开在线demo的角色信息设计吗? HOT 1
- 请问未来是否可开发web demo和api的新功能呢? HOT 1
- [Question]: Clarification on the division of full user dialogue history into context and history HOT 4
- [Bug]:关于训练阶段数据预处理逻辑的问题 HOT 1
- [Question]: HOT 1
- [Question]:
- [Question]:请问persona_benchmark.json的格式是什么样 HOT 1
- [Feature]:数据开源或者数据获取方式? HOT 3
- 论文写的实在是太不清晰了 HOT 1
- bug report
- [Question]:How to train custom character data HOT 1
- [Bug]:The model will return wrong or duplicate data HOT 2
- [Question]:配置完毕以后,执行的时候出这个问题,不知道是不是少文件了 HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from chatplug.