messages can be sent to gpt and messages are received.
documents are uploaded when the message is sent. To stream line this and reduce cost we will need to "fine-tune" the gpt.
to do this we will need to come up with training statements and store them in a file that is uploaded to the open ai api.
Our next steps.
put the chat system behind a user login.
put it on a testing server.
verify that this process is okay with open ai.
look at the instruct series.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Putting chat system behind login:
User is passed through workflow.
instead we need to make a database to store the messages.
the user id should be a relationship on the messages.
The user should have an obfuscated private channel assigned to them.
We need to make the table for messages next.
Messages are not being received on private channels
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Create a job that sends messages to GPT-3
To send a message set up a client
request an answer
Upload document to the GPT-3
messages can be sent to gpt and messages are received.
documents are uploaded when the message is sent. To stream line this and reduce cost we will need to "fine-tune" the gpt.
to do this we will need to come up with training statements and store them in a file that is uploaded to the open ai api.
Our next steps.
put the chat system behind a user login.
put it on a testing server.
verify that this process is okay with open ai.
look at the instruct series.
Putting chat system behind login:
User is passed through workflow.
instead we need to make a database to store the messages.
the user id should be a relationship on the messages.
The user should have an obfuscated private channel assigned to them.
We need to make the table for messages next.
Messages are not being received on private channels