geospatial enviornment python 3.12

A geospatial Python environment for working with raster data, vector data, satellite imagery, and machine learning. Built on top of GeoPandas, Rasterio, PyTorch, and JupyterLab.

---

Requirements

Platform	Python	File
macOS (Apple Silicon / Intel)	3.12	reqtext_p312_mac.txt
Linux x86_64	3.12	reqtext_p312_linux.txt

Note: The two files differ because some packages (pyobjc, appnope) are macOS-only, and several packages like rasterio 1.5.0 and torch 2.12.0 require Python 3.12 which is only used on the Linux side.

---

Prerequisites

macOS

1. Homebrew — if not already installed:

   /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

2. GDAL — required by rasterio:

   brew install gdal

3. uv — fast Python package manager:

   brew install uv

Linux (Ubuntu / Debian)

1. GDAL — required by rasterio:

   sudo apt update
   sudo apt install gdal-bin libgdal-dev python3-dev build-essential

2. uv — fast Python package manager:

   curl -LsSf https://astral.sh/uv/install.sh | sh

---

Installation

macOS

# 1. Clone the repo
git clone https://github.com/thescriptingcave/geospatial.git
cd geospatial

# 2. Create and activate a virtual environment with Python 3.11
uv venv --python 3.12
source .venv/bin/activate

# 3. Install dependencies
uv pip install -r reqtext_p312_mac.txt

Linux

# 1. Clone the repo
git clone https://github.com/thescriptingcave/geospatial.git
cd geospatial

# 2. Install Python 3.12 if not already available
uv python install 3.12

# 3. Create and activate a virtual environment with Python 3.12
uv venv --python 3.12
source .venv/bin/activate

# 4. Install dependencies
uv pip install -r reqtext_p312_linux.txt

---

Starting JupyterLab

Once your environment is activated:

jupyter lab

This will open JupyterLab in your browser at http://localhost:8888.

---

Key Packages

Category	Packages
Geospatial	GeoPandas, Rasterio, Shapely, PyProj, osmnx, Folium
Raster / imagery	Rasterio, imagecodecs, tifffile, imageio, opencv-headless
Machine learning	PyTorch, torchvision, scikit-learn, scikit-image
Data science	NumPy, SciPy, Pandas, Matplotlib, SymPy
Notebooks	JupyterLab, ipywidgets, nbconvert

---

Troubleshooting

rasterio fails to import on Linux Make sure GDAL is installed at the system level:

sudo apt install gdal-bin libgdal-dev

uv can't find Python 3.12 on Linux Let uv download and manage it:

uv python install 3.12

pyobjc errors on Linux This package is macOS-only and is not included in reqtext_p312_linux.txt. Make sure you're using the correct requirements file for your platform.

Dependency conflicts Always install inside a fresh virtual environment. If you hit conflicts, try:

deactivate
rm -rf .venv
uv venv --python 3.12   # or 3.11 on macOS
source .venv/bin/activate
uv pip install -r reqtext_p312_linux.txt

---

Project Structure

geo_env/
├── reqtext_p312_mac.txt       # macOS dependencies (Python 3.11)
├── reqtext_p312_linux.txt     # Linux x86_64 dependencies (Python 3.12)
├── README.md
└── notebooks/                 # Jupyter notebooks go here

---

License

MIT