John Horton comments

Results 150 comments of


                                            John Horton

For each inference service, add methods to learn tokens-per-minute and requests-per-minute limits

@zer0dss This naturally goes w/ price look-ups.

Add notion of "sections" to Survey object; allow section-specific skip-logic; Create Survey CSS

Notes of creating a custom CSS: https://chat.openai.com/share/38dbd588-dc00-4415-be22-59b4e631798f

Add notion of "sections" to Survey object; allow section-specific skip-logic; Create Survey CSS

@onmyraedar / @apostolosfilippas This issue is related to my "simualted_results" branch yesterday. If we have human survey taking, we'll want to expand the skip-logic to the notion of question groups...

Results `sort_by()` method does not behave as expected

Can you say a bit more here - what's the expected behavior?

Results `sort_by()` method does not behave as expected

I see. This might be challenging because it would require the dataset to have a "memory" of what is sorted by and I'd prefer to keep it stateless. What about...

Results `sort_by()` method does not behave as expected

New version does sort this way - can you update docs, @rbyh ?

Results objects should have add method

They do! ```python Results.example() + results.example() ```

Results objects should have add method

@rbyh Can you document?

Object .example() method may be confusing

Agreed - two thoughts. 1. We should encourage people to use help(X) to get our docs in context (and include URL to docs) 2. If we get the 'code' method...

Object .example() method may be confusing

I'm going to close this and suggest that we focus on educating users about 1 and making 2 'work'