plotly_cheatsheet.pdf: a collection of Plotly visualization preferences.vis_guide.md: a markdown file that serves as a guide for LLMs to generate visualizations using Plotly.
You can use vis_guide.md in two ways:
- Copy the content of the file to
.cursorrules. This will provide visualization context for LLMs without the need to manually instruct it every session. - Tag the file with
@vis_guide.mdin your prompt. This will provide the LLM with the context of the file for that specific session. This is outlined in the example below.
Note: edit the file to customize it for your specific needs and preferences, for example the Limitations section contains hard-coded values that you may want to change.
Write a Python script to perform Exploratory Data Analysis (EDA) on the Titanic dataset (`data/train.csv`).
- The target variable is `Survived`.
- The EDA should be **visualization-heavy**: generate as many meaningful plots as possible to help understand relationships between features and the target variable.
- Each plot should be saved to a dedicated `plots/` folder (create it if it doesn't exist).
- The script should follow the structure and recommendations of @vis_guide.md
- Include univariate, bivariate, and multivariate visualizations where appropriate.
- Include print statements summarizing insights after each plot.
eda.py: Python script that performs basic visual EDA on the Titanic dataset.plots/: Directory containing all the plots generated by the script.