Hi all,
has anyone used GPT3 finetuning for multilabel text classification?
I would highly appreciate to see an example how to structure the completion and learn from best practices.
For clarification:
My ask is not about the common (multiclass) classification use case.
Multilabel means that each instance has 1 or more labels associated.
Thank you
Andreas
@akopp Find any good resources here? Iâm poking around myself. If I find anything, Iâll post it here as well.
In my case I was able to achieve good results using entity extraction similar to this example: OpenAI API
Not sure if this is generally a good approach for multilabel but in my scenario it worked quite well.
I donât see why you wouldnât be able to use their finetune classification example for multiclass⌠You would just feed multiple classes into the training data, right? Give it a go!
I am also interested in how we can fine-tune the model for multi-label classification. If anyone has figured it out let me know please or if there is any notebook on how to do that?
@saraahmadi321 @akopp @aaron5 have you guys found any solutions? I want to do something similar where i want to multi-label customers feedback. Thanks in advance!
You can create a fine-tune by creating a jsonl file with example prompts and completions. So a file with hundreds or thousands of lines like this:
{âpromptâ: âYour company is awesome.\n\n###\n\nâ, âcompletionâ: " 1"}
{âpromptâ: âYour company is bad.\n\n###\n\nâ, âcompletionâ: " 0"}
Note the space before the completion output. Also, the '\n\n###\n\n" is what I used as a stop word, which is the same one recommended by OpenAI.
Then to create the fine-tune, follow along with the docs:
https://beta.openai.com/docs/guides/fine-tuning
Here would then be the parameters you would send when you call your fine-tuned model. The important ones being the temperature of 0, and in my case, since I am just outputting single character labels, I set max_tokens to 1.
{âtemperatureâ: 0, âmax_tokensâ: 1, âtop_pâ: 1, âlogprobsâ: 2, âfrequency_penaltyâ: 0, âpresence_penaltyâ: 0}
Be sure to append the stop word of â\n\n###\n\nâ to the input prior to sending it off to the trained fine-tune.
Note: You can only fine-tune base models right now. So âdavinciâ and not âtext-davinci-003â.
For fine-tunes with a lot of training data, you can get away with using ada or babbage for classification. But you can train various versions and see which ones work better for you. The higher end models such as curie or davinci would be used if you have limited training data.
Thatâs basically the high level advice on creating a fine-tune.
Good luck!
This is great advice. I was playing around with fine-tuning as well for this. If you limit the max tokens to 1 though isnât that still a multi-class instead of a multi-label model? Even with fine-tuning Iâve been struggling to get the model to spit out the correct set of labels using the TRAINING data⌠sigh.
Which model did you fine-tune?
How many prompt/completion pairs did you provide it for training?
Did you include the stop token?
The classification would map to single character output so, â0â, â1â, ⌠âaâ, âbâ, ⌠âzâ
Then you map this to what the meaning is.
To map one thing to two or more different categories, you need to run two or more models against the same input. But the same idea.
Oooooooooh I think I see what you mean. I can map a single character to multiple labels. Thatâs clever! The data-prep CLI tool recommended I use ada so thatâs what I went with. I did NOT include the stop token (which I will now do). I fed it ~200 prompts with 5 labels/completions. Even with the temp at 0 I was getting some wild outputs.
With 200 training points, you might want to bump up to babbage or curie (and davinci as a last resort). The higher models might be able to âlearn quickerâ since they have more parameters.
Otherwise beef up your training data if you want to use ada and the stop token doesnât fix it.
Hi Curt,
I am doing multiclass classification with 230 labels and 120k training samples. What model would you recommend?
I would start with Babbage and go up if the results arenât good. But that is a ton of labels. Maybe break it up into several course classifier groups up front and further refine with additional classifiers.
@dml1002313 Did you solve your issue. Iâm looking to perform multi-label classification of text into 7 labels. None, one or many labels can apply to the text, I already have my dataset in the format : âthis is a sampleâ,0,1,0,1,1,0,0,1 and trying to work out how best to structure my dataset for fine-tuning GPT3 so that I can get a response such as :
prompt: this is an example test
completion: 1,1,1,0,1,0,0,0
@aaron5 @jerome1
hello, have you guys found any solutions? I want to do something similar where i want to multi-label classification. Thanks in advance!
@maysaa.khalil Yes. But I also forget the exact context I was thinking about when I responded up top lol. Can you do a super quick summary of your use case?
Hi @akansel thanks a lot for your reply, it is the idea or solution on how to fine-tune a multi-label classification problem with openai models. Thanks again 
@maysaa.khalil Weâve almost entirely migrated away from using fine-tuned models for classification in favor of 3.5 turbo. Itâs cheaper and faster and when you combine it with validation I think itâs more effective. On that same note, historically weâve used fine-tuned models to perform the validation check, but even in that case weâre evaluating using 3.5 turbo instead.
To speak specifically to multi-label classification without knowing the specifics of your use case might be tricky, but overall I would recommend putting formatting expectations into your prompts in a very specific way. Iâve found that clearly defining the expected prefixes, suffixes and delimiters not only allows you to generate cleaner variables for your code, but the models also perform multi-label classifications more effectively when you designate the format. Sort of like youâre defining multiple âcontainersâ of generated content in the same request. This makes it more obvious to you AND the model when a particular container is empty.
For suggestions on good formatting instructions, Iâd recommend GPT-4 via chat. Lay out your use case and the classification decisions youâd like the model to return, and ask it to suggest formatting guidelines that can be used to clearly differentiate the various aspects of the request.
If youâre using 3.5 turbo instead of fine-tuning a model, are you not providing hundreds of user validated examples to refine the results youâre getting?
@devbydylan The advantage of this approach is that you donât have to. Iâll give you an example:
Letâs say youâre evaluating user reviews and you want to assign each review one value called âhappy_or_madâ to indicate sentiment and one value called âpurchased_itemâ that tries to identify what the user actually bought. Rather than feed it hundreds of validated reviews, you engineer a prompt the best that you can, then run it hundreds of times on real data or generated data. After reviewing the results, you make adjustments to the prompt to account for cases where it got one or more of the values wrong. If you canât get it to the desired degree of accuracy after doing this a couple times, you create a separate prompt for the review model to look out specifically for those tricky reviews that your first model canât seem to nail. Because this second prompt is focused exclusively on exceptional use cases, it shouldnât have any problems identifying and correcting them. The advantage of GPT 3.5 is it super fast and cheap so you can afford to run each review twice. Makes sense? If you want to share your use case or something similar Iâd be happy to help you think through it.