You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I just used Pythagora via its VsCode extension. I know that I would need an api key for LLM and that openai api key is recommended, for that, I prepared an openai api key.
I thought that when I start using Pythagora for the first time, it will ask me: "what type of LLM do I want to use?", then it will create the configuration file (config.json) that corresponds to my choices for the LLM then it opens the config file asking me to provide the api keys.
in addition to that the config file was not at the same level as the workspace folder, so it is not visible in vscode directly.
the configuration of Pythagoras must be more intuitive for a smoother and more pleasant user experience.
In my opinion it must have Pythagora-Installer a program or executable whose role is:
create a folder (if it does not exist) .Pythagora ( at the root of the current user directory, something like /home/.pythagora
create a folder with a path similar to the absolute path of the current folder: for example if the vscode extension is enabled for a folder with a path /home/document/proj1 we create the folder .pythagora/home/document/proj1
then create a virtual python environment in the folder created in 2)
activate the virtual environment created in 3)
Install the python package GPT-PILOT using 4)
so we do not need to ask the user any questions before he can use pythagora except the type of LLM he is going to use and provide these keys in the configuration file which will be opened before him
The text was updated successfully, but these errors were encountered:
Version
VisualStudio Code extension
Suggestion
I just used Pythagora via its VsCode extension. I know that I would need an api key for LLM and that openai api key is recommended, for that, I prepared an openai api key.
I thought that when I start using Pythagora for the first time, it will ask me: "what type of LLM do I want to use?", then it will create the configuration file (config.json) that corresponds to my choices for the LLM then it opens the config file asking me to provide the api keys.
in addition to that the config file was not at the same level as the workspace folder, so it is not visible in vscode directly.
the configuration of Pythagoras must be more intuitive for a smoother and more pleasant user experience.
In my opinion it must have Pythagora-Installer a program or executable whose role is:
create a folder (if it does not exist) .Pythagora ( at the root of the current user directory, something like /home/.pythagora
create a folder with a path similar to the absolute path of the current folder: for example if the vscode extension is enabled for a folder with a path /home/document/proj1 we create the folder .pythagora/home/document/proj1
then create a virtual python environment in the folder created in 2)
activate the virtual environment created in 3)
Install the python package GPT-PILOT using 4)
so we do not need to ask the user any questions before he can use pythagora except the type of LLM he is going to use and provide these keys in the configuration file which will be opened before him
The text was updated successfully, but these errors were encountered: