introduction to Python for the Casey Lab

the purpose of this tutorial is to help you get started with Python.

when to consider switching

• large datasets
• spatial analysis
• slow r code
• large rasters and satellite imagery
• Google Earth Engine
• parallel processing

contents

• 01_python_to_r/python_to_r.Rmd - R Markdown file with Python code to translate (no answers)
• 01_python_to_r/python_to_r_with_answers.Rmd - R Markdown file with answers
• 02_python_practice/python_practice.ipynb - Jupyter notebook for practicing Python basics
• 02_python_practice/python_practice_with_answers.ipynb - Jupyter notebook with answers to the practice problems
• 03_python_spatial/spatial_data.ipynb - Jupyter notebook for practicing spatial data analysis

---

python setup

1. clone this repository

2. install vscode

see 00_setup/02_vscode_setup.md for instructions.

3. set up python and environments

see 00_setup/01_python_setup.md for instructions.

4. For the future, set up a pre-commit hook

https://pre-commit.com/ ensures that you do not push a jupyter notebook to github with output in it. this is important because you may mistakenly push something that should not be on github, or that is too big to be on github. the pre-commit hook will check for output in your jupyter notebooks. if there is output, it will fail, then it will clear the output, then you can add, commit, and push again.

to set this up:

1. pip install pre-commit
2. pre-commit install
3. make sure the .pre-commit-config.yaml file is in the root of your repository. you can copy and paste this exact file into each of your repositories.

---

tutorial exercises

exercise 1: python to r translation

practice translating Python code to R using the python_to_r.Rmd file.

instructions:

1. open 01_python_to_r/python_to_r.Rmd in VSCode
2. read through the Python code
3. write the equivalent R code in the empty chunk
4. check your work against 01_python_to_r/python_to_r_with_answers.Rmd

exercise 2: python basics notebook

practice python fundamentals with the jupyter notebook.

instructions:

1. make sure your environment is activated: conda activate practice_env
2. open 02_python_practice/python_practice.ipynb in VSCode
3. work through the exercises, running each cell with Shift+Enter
4. complete the practice problems at the end

exercise 3: spatial data analysis

practice spatial data analysis with the jupyter notebook.

instructions:

1. make sure your environment is activated: conda activate practice_env
2. open 03_python_spatial/spatial_data.ipynb in VSCode
3. work through the exercises, running each cell with Shift+Enter
4. complete the practice problems at the end

---

quick reference: python vs r

task	python (pandas)	r (tidyverse)
read CSV	pd.read_csv("file.csv")	read_csv("file.csv")
view head	df.head()	head(df)
column mean	df['col'].mean()	mean(df$col)
filter rows	df[df['col'] > 5]	filter(df, col > 5)
select columns	df[['a', 'b']]	select(df, a, b)
rename column	df.rename(columns={'old': 'new'})	rename(df, new = old)
new column	df['new'] = df['a'] + df['b']	mutate(df, new = a + b)

---

tips

• indentation matters in Python! use 4 spaces (not tabs) for code blocks
• 0-indexed: python counts from 0, R counts from 1
• assignment: Python uses =, R uses <- (though = also works)
• use print(): unlike R, Python won't automatically display results in scripts

---

troubleshooting

"module not found" error:

pip install module_name

conda environment issues:

conda deactivate
conda activate practice_env

jupyter can't find kernel:

pip install ipykernel
python -m ipykernel install --user --name=practice_env