Not getting results as aspected?

[alan-ai-learner]

@abhijithneilabraham thanks a lot for your beautiful work, i have some questions,

1. What is the efficient way to create the schema, like we cannot analyze the data every time, and create the schema for that?
2. I tried with some custom datasets its performing well but after passing the schema , it is not recognizing some of the columns?
3. Can I fine tune your model?

Thanks in advance!

[abhijithneilabraham]

1. Schema can be created automatically too. There's a function get_schema_for_csv inside data_utils.py that performs this operation.
2. Could you give me more info on this?
3. This is not a pretrained model, and rather is an AI assisted tool which uses pre trained models under the hood. Maybe you can fine tune those pretrained models and evaluate on top of that. Here's a blog which can help you know more about how TableQA works.

Thanks