hello, i want to fine-tune an llm using sentiment analysis, and as far as i understand, i need to generate a synthetic dataset for this.

actually, i already have around 1000 pieces of data related to a product. i can directly ask the llm to perform sentiment analysis on these, but the llm struggles to make clear inferences for some comments.

what exact steps should i follow?

hello, i want to get multiple structured output from llm using llamaindex. for example when i query to llm, then the answer should like:
query: List the most populous cities in England.



How can do this?

@kapa.ai how can I solve this problem?