AI Agent Coding in Positron - Can it be done? скачать в хорошем качестве

AI Agent Coding in Positron - Can it be done?

In this video I put the new Positron Data Science IDE from Posit to the test. Can I use an agentic workflow to analyze a dataset using R in a Quarto Document? Short answer: Yes. But at a cost. The data set is available in GitHub at https://github.com/juanklopper/Tutori... The agent instruction is: --- applyTo: "quarto" --- Quarto Instructions Start the document with a YAML header. Use `title`, `author`, and `date` in the YAML header. ``` title: "Your Document Title" author: "Your Name" date: "`r Sys.Date()`" format: html execute: echo: true warning: false error: false ``` Use the R language for code chunks. Use markdown for content. Create sections with `##` for main headings and `###` for subheadings. Use `-` for bullet points and `1.` for numbered lists. Use backticks for inline code and triple backticks for code blocks. Use `` for images. Interpret the results of all analysis and provide insights. Use ggplot2 for data visualizations. The prompt is: The "Metrics.CSV" file in this folder contains data on an experiment to determine the sensitivity, specificity, and the positive and negative predictive values of a new test for a specific disease. A total of 322 participants were selected for the experiment. To reflect the current prevalence of the disease in the community, which is 18.6%, 60 participants have the disease and 262 do not. The `Disease` column has two classes. `No` indicates the absence of the disease as measured by a gold-standard test, and `Yes` indicates the presence of the disease by the same gold-standard test. The `Test` column has two classes and records the results of the new test. `Negative` indicates a negative test for the disease and `Positive` indicates a positive rest for the disease. Analyze the data using the following plan: 1. Calculate and visualize the frequency and relative frequency of each class in each of the `Disease` and the `Test` columns. 2. Create a contingency table of the two columns and visualize the results. 3. Calculate and interpret the joint probabilities of the contingency table. 4. Calculate and interpret the marginal probabilities of contingency table. 5. Calculate and interpret the sensitivity and the specificity of the new test. 6. Calculate and interpret the positive and the negative predictive values of the new test. 7. Recalculate the positive predictive value and the negative predictive value for a prevalence of 40% using the law of total probability. Explain why the increased prevalence changes the positive predictive value and the negative predictive value. 8. Write a summary and conclusion of the analysis of the new test.