Help in classification a database

I have a database with several hashtags extracted from Instagram, where I need to classify what is an accident and what is not. I can’t upload my base, I think my script is broken. Is it a master’s work, someone to collaborate or help me with?
My language is Portuguese, sorry if something wasn’t clear.

can you give some examples of what is an accident and what isn’t? You will probably need to combine some SQL queries with Python to check.

Yes, I can, I have a series of tweets that inform me that there was an accident with Portuguese caravels and not others.

In my json file I have these database examples:

{“text”: “Beautifully dangerous \n#caravel #caravelaportuguesa #aracaju #aracajusergipe”, “label”: “NoAccident”}
{“text”: “God is extravagant in his creation! \nP.s Too bad I felt the strength of this caravel in my skin! Rs\n\n#CaravelaPortuguesa #Caravela #Cnidario #Nature #Nature #Paiva #Beach", “label”: “YesAccident”}

I have a base that I know what is an accident and what is not, but I would like to see OpenAI rate for me, so I left the first 5 defined as “NoAccident” or “YesAccident” and the rest I left blank, as per example:

{“text”: “#caravelaportuguesa”, “label”: “”}

In json itself (I don’t know if that would be right).

So I made the script to upload my file and I made my script to run the sort, but I’m getting the following error:

openai.error.InvalidRequestError: [[‘base-test’]] is not of type ‘string’ - ‘file’

I’ve already searched for solutions on the Internet, but I haven’t been able to solve it yet. If you can help me, I would greatly appreciate it!! :slight_smile: