How to visualize a Pandas Dataframe in R chunk Quarto?

Question

I was wondering if there is an option to use a Pandas dataframe created in a Python chunk, in your R chunk to visualize the dataframe. Here is some reproducible code:

---
title: "How to use Pandas Dataframe in R chunk"
format: html
engine: knitr
---

Example code:

```{python}
import pandas as pd
df = pd.DataFrame(data = [[1,2],[2,3],[3,3]], columns = ['v1', 'v2'])
print(df)
```

```{r}
library(ggplot2)
ggplot(data = df, aes(x = v1, y = v2)) + geom_point()
```

This produces the following error:

Error in `ggplot()`:
!   You're passing a function as global data.
  Have you misspelled the `data` argument in `ggplot()`
Backtrace:
 1. ggplot2::ggplot(data = df, aes(x = v1, y = v2))
 2. ggplot2:::ggplot.function(data = df, aes(x = v1, y = v2))

This happens of course because it can’t find the dataframe df created in the python chunk. I also loaded the reticulate package and tried using py$df in R chunk without success. So I was wondering if anyone knows how to use a Pandas dataframe in your R chunk to visualize it in Quarto?

Asked By: Quinten

||

Source

Answer 1

From the {reticulate} docs,

All objects created within Python chunks are available to R using the py object exported by the reticulate package.

which means that you have to either use library(reticulate) in the qmd file or at least use reticulate::py

---
title: "How to use Pandas Dataframe in R chunk"
format: html
engine: knitr
---

Example code:

```{python}
import pandas as pd
df = pd.DataFrame(data = [[1,2],[2,3],[3,3]], columns = ['v1', 'v2'])
print(df)
```

```{r}
library(ggplot2)

ggplot(data = reticulate::py$df, aes(x = v1, y = v2)) + geom_point()
```


```{r}
library(reticulate)

ggplot(data = py$df, aes(x = v1, y = v2)) + geom_point()
```

Answered By: shafee

How to visualize a Pandas Dataframe in R chunk Quarto?

Question:

Answers: