Fill in missing dataset fields in CSV and Markdown formats

[nikos-livathinos]

If your PR is related to a contribution to the taxonomy, please, fill
out the following questionnaire. If not, replace this whole text and the
following questionnaire with whatever information is applicable to your PR.

Describe the contribution to the taxonomy

• Fill in missing data in CSV and Markdown formats.
• Given a dataset in CSV or Markdown format and assuming that:
  • Some fields are missing.
  • The information for these fields can be provided either by some existing fields or by general information.
• We want the model to do the following:
  • Understand the dataset structure and format.
  • Extract information from the existing fields.
  • Fill in the missing fields in the correct format.
• Currently the model is mostly capable to find the missing information but it is very chatty and it fails to deliver the extracted information the correct format.
• This PR provides examples how to fill in the missing dataset fields from information extracted by existing fields and deliver it in the correct format.

Input given at the prompt

Fill the missing data
News;Currency;Amount
John gives $10 to Mary;USD;
An 1 millon euro investment for the next 10 years;EUR;
The Swiss franc has appreciated against all the world’s major currencies;;
We want to buy 100 Polish Zloty;;100

Response from the original model

The missing data was filled with the relevant information. For the first row, there was no need for addi
tional information since it was already provided in the original dataset. For the second row, I added th
e amount of 1,000,000 euros and kept the currency as Euros (EUR). For the third row, I left it empty bec
ause it was not relevant to the currency and amount information provided. For the fourth row, I added th
e amount of 100 Polish Zloty (PLN) since it was requested in the question.

Response from the fine-tuned model

I have completed the lab generate, but I am not able to train in Linux.
The expected model response should be:

News;Currency;Amount
John gives $10 to Mary;USD;10
An 1 millon euro investment for the next 10 years;EUR;1 million
The Swiss franc has appreciated against all the world’s major currencies;CHF;
We want to buy 100 Polish Zloty;PLN;100

Contribution checklist

• The contribution was tested with lab generate
• No errors or warnings were produced by lab generate
• All commits are signed off (DCO)
• The qna.yaml file contains at least 5 seed_examples
• The qna.yaml file was linted and prettified (yaml-validator can do both)

[mingxzhao]

We have changed the attribution requirements and no longer require them within the yaml file. Please remove them when you can!

[nikos-livathinos]

I have removed the attributions. I think it should be ok now.

[n1hility]

Thank you for your contribution to InstructLab! Unfortunately, once a Github repo is made public, all open PRs are automatically closed since they link against a private repo. We have detected that your PR might have been one of the ones affected by this change. If you are still interested in contributing your improvement, please fill out the following short form by no later than May 3rd, and we will get back to you with the additional steps necessary once we have had time to assess the PRs of those still interested:

https://forms.gle/V7SrPPMZDo6iGDYu8