python_rmd.Rmd

---
title: "Python in R Markdown"
output: 
  github_document:
    toc: true
authors: "Tiffany Timbers and modified by Kewalin Samart" 
editor_options: 
  markdown: 
    wrap: 72
---

To use Python in R Markdown, we need to load the
{[reticulate](https://rstudio.github.io/reticulate/)} R package in a
code chunk, we do it here in the setup code chunk. We also load the
{[ggplot2](https://ggplot2.tidyverse.org/)} R package so we can plot
some data from a Python `pandas` data frame later on!

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(reticulate)
library(ggplot2)
```

## Python in `.Rmd` files

### Load libraries

Just like R, you can use Python in `.Rmd` files! Here we import the
Python package that we need to import our data into Python, `pandas`.
`pandas` is a Python package that adds data reading, wrangling, and
simple data visualization functionality to Python where sometimes depend
on `matplotlib` functions. It holds a similar place as the {tidyverse} R
meta-package does (however `pandas` is not a meta-package, just a very
large package). If you want to learn more about pandas, see the "[10
minutes to
pandas](https://pandas.pydata.org/pandas-docs/stable/user_guide/10min.html)"
in the docs our checkout this free interactive course:
<https://prog-learn.mds.ubc.ca/en>.

To [install Python
packages](https://rstudio.github.io/reticulate/articles/python_packages.html)
please run the cell below

```{r}
py_config() # optional: to check python configuration setups
py_install("pandas") # install pandas
py_install("matplotlib") # install matplotlib
```

### Downgrade pandas and numpy version

#### to make later python\<-\>R conversions possible

ref: <https://github.com/rstudio/reticulate/issues/1620>

```{r downgrade}
reticulate::py_install(c("pandas<2.2.2", "numpy<2.0"), ignore_installed = TRUE)
```

```{python libraries}
import pandas as pd
import matplotlib.pyplot as plt
```

### Load data

Let's load the titanic data (which lives in the `data` directory of this
project) and view the data:

```{python data}
titanic  = pd.read_csv("data/titanic.csv")
```

### plot data

#### Using Python and `pandas`

First, let's use some quick and simple `pandas` plotting functions:

```{python python-plot}
titanic.plot.scatter(x='age', y='fare', alpha=0.3)
plt.show()
```

Want to learn more about getting started plotting in Python using
`pandas`, see this page to get started:
<https://pandas.pydata.org/pandas-docs/stable/user_guide/visualization.html>

#### Using R and {ggplot2}

We can also access the Python environment in this R Markdown document
from R! This allows us to apply R's functions on our Python objects! We
can use the `py$obbject` syntax to do this and create a data
visualization using {ggplot2}!

```{r r-plot, warning=FALSE, fig.height=4, fig.width=4}
ggplot2::ggplot(py$titanic, aes(x = age, y = fare)) +
  geom_point()
```

### Subsetting data in `pandas`

Here we subset the age and fare columns:

```{python}
titanic[["age", "fare"]]
```

Here we filter for rows containing passengers who's age was 70 or
greater:

```{python}
titanic[titanic["age"] > 70]
```

Here we find the destination of the first passenger:

```{python first destination}
first_dest = titanic["home.dest"][0]
first_dest
```

Why 0 as the numerical index above? Python counts from 0, not 1!

### And inline python code within markdown text!

We can use the R Markdown inline syntax and the `py$` syntax to reach
into the Python environment and access the `first_dest` python object
(which is a string).

The destination of the first passenger is `r py$first_dest`.

### Sourcing a Python script

#### Example 1: `greeting.py` (contains a function named greeting with Python code to call the function within the same script)

```{r}
# source the Python script
source_python('greeting.py')

# call the Python function and pass a new name
greeting("Kewalin")
```

#### Example 2: `titanic.py` (only contains a function named read_titanic)

```{r}
# source the Python script
source_python("titanic.py")

# call the Python function and pass the Titanic dataset
titanic <- read_titanic("data/titanic.csv")

# load ggplot2 for visualization
library(ggplot2)

# plot the Titanic data
ggplot(titanic, aes(x = age, y = fare, color = as.factor(survived))) +
  geom_point(alpha = 0.6) +
  labs(
    title = "Age vs Fare (Colored by Survival)",
    x = "Age",
    y = "Fare",
    color = "Survived"
  ) +
  theme_minimal()

```