serp-ai / chatgpt-plugins Goto Github PK
View Code? Open in Web Editor NEWRepo for giving ChatGPT the ability to use web browsing, python code execution, and custom plugins
Home Page: https://serp.ai/tools/chatgpt-plugins/
Repo for giving ChatGPT the ability to use web browsing, python code execution, and custom plugins
Home Page: https://serp.ai/tools/chatgpt-plugins/
You guys need to improve the installation instruction. It's not clear what is meant by "Run all of the setup cells"
You should also set up the API KEY as a definition that can be defined at the top of the plugins.ipynb file
It's not clear at all how to get this up and running
Whenever I turn on short term memory I get the following error:
Thought: I need to find a reliable source for AI news.
Action: Web Search
Action Input: "latest AI news"
Searching for: latest AI news...
Error communicating with chatGPT: expected string or buffer
Error communicating with chatGPT: expected string or buffer
Output exceeds the [size limit](command:workbench.action.openSettings?%5B%22notebook.output.textLineLimit%22%5D). Open the full output data [in a text editor](command:workbench.action.openLargeOutput?073c5e0f-65ab-48f9-ab3b-c7000d090f1f)---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
Cell In[29], line 2
1 # Example
----> 2 latest_news_summary, cache = action_loop(assistant, user_prompt="""Find the latest in ai news and then give me a 500-word summary. No python code allowed.
3 Final Answer must be in the following format:
4 Final Answer:
...
5
...
6
...
7
...
8 ...""", function_definitions=function_definitions, cache=cache, temperature=0.9)
Cell In[26], line 47, in action_loop(assistant, user_prompt, cache, function_definitions, max_observation_tokens, memory_tokens, max_loops, sleep, **kwargs)
45 break
46 # get the next action
---> 47 response = assistant.get_chat_response(prompt=None, stop=stop, inject_messages=[{0: {'role': 'user', 'content': user_prompt}}, {1: {'role': 'assistant', 'content': next_message}}], **kwargs)
49 clear_output(wait=True)
50 print(next_message + response.choices[0].message.content.strip())
File [c:\Users\User\\ChatGPT-Plugins-main\assistant.py:401](file:///C:/Users/User/ChatGPT-Plugins-main/assistant.py:401), in OpenAIAssistant.get_chat_response(self, prompt, max_tokens, temperature, top_p, n, stream, frequency_penalty, presence_penalty, stop, logit_bias, user, max_retries, inject_messages)
399 iteration += 1
400 if iteration >= max_retries:
--> 401 raise e
402 print('Error communicating with chatGPT:', e)
403 sleep(1)
...
--> 116 if match := _special_token_regex(disallowed_special).search(text):
117 raise_disallowed_special_token(match.group())
119 try:
TypeError: expected string or buffer/
I followed the instructions and at the run Setup, I get the following error:
TypeError Traceback (most recent call last)
/var/folders/wm/c0rmwv6x5rn95tg5bh4r71hr0000gn/T/ipykernel_36563/1860396667.py in
7 from IPython.display import clear_output
8
----> 9 from tools import WebSearch, WebBrowser, python_runner
10 from assistant import OpenAIAssistant
~/Desktop/ChatGPT-Plugins-main/tools.py in
111
112
--> 113 class WebSearch():
114 """
115 Class to handle web search
~/Desktop/ChatGPT-Plugins-main/tools.py in WebSearch()
122 self.cache = {}
123
--> 124 def search(self, keywords: Any, region: str = "us-en", safesearch: str = "Off", time: Any | None = 'y', max_results: Any | None = 20, page: int = 1, output: Any | None = None, download: bool = False, cache: bool = False) -> list:
125 """
126 Search
TypeError: unsupported operand type(s) for |: '_SpecialForm' and 'NoneType'
was looking to try something like this but i would rather use the chat models since they're better in multiple aspects
however i cant really get them to follow the format, is it possible to get them to work for this?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.