Update BuildChartPrompt.txt with Templates #1244
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
New Chart Template Prompt with sample templates Jinja2 instead of fstrings for returning HTML Other Altair and HTML Specific Guidelines
llms/BuildChartPrompt.txt
Outdated
| housing = load_data(path) | ||
|
|
||
| # Inspect schema (useful for debugging) | ||
| print("=== Column data types ===") |
Collaborator
There was a problem hiding this comment.
Do we need this? This takes up context space and I don't think adds anything to charting?
Collaborator
Author
There was a problem hiding this comment.
reverted to the earlier example, as the templates cover this. we need this as sometimes the AI Might get stuck in a loop of "Fix" and it cannot determine the datatype of the columns. This added context helps it to "Fix" it in the next run.
| Data Handling Requirements: | ||
| - When working with CSV files from external sources, always include encoding handling: Use encoding="latin1" or encoding="utf-8-sig" for broader compatibility | ||
| - Always clean column names (strip whitespace, handle special characters) before using them in data visualizations or chart encodings | ||
| - NEVER ASSUME THE SCHEMA OF THE DATASET. Always create UDF to look at the sample of the dataset to find it out if you are not sure about the schema |
Collaborator
There was a problem hiding this comment.
This is not too helpful because every UDF has only a single run. Should reframe this on how UDF can get content of tables by getting context from another UDF
Collaborator
Author
There was a problem hiding this comment.
The behaviour I noticed which led me to add this was that
- For some partially known/famous dataset, it directly jumps on creating a chart udf, Assuming the column names itself which lead to a lot of errors. We need to think about what happens after we click on "Fix" if and when needed, does it get enough context to debug or not.
- Can remove the clean column directive - Agreed
- For CSV Files from a lot of sources, it was giving me encoding errors while reading csv directly without the encoding parameter, which led me to adding this line.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
New Chart Template Prompt with sample templates
Jinja2 instead of fstrings for returning HTML
Other Altair and HTML Specific Guidelines