CARTOframes

A Python package for integrating CARTO maps, analysis, and data services into data science workflows.

Python data analysis workflows often rely on the de facto standards pandas and Jupyter notebooks. Integrating CARTO into this workflow saves data scientists time and energy by not having to export datasets as files or retain multiple copies of the data. Instead, CARTOframes give the ability to communicate reproducible analysis while providing the ability to gain from CARTO’s services like hosted, dynamic or static maps and Data Observatory augmentation.

Features

• Write pandas DataFrames to CARTO tables
• Read CARTO tables and queries into pandas DataFrames
• Create customizable, interactive CARTO maps in a Jupyter notebook
• Interact with CARTO’s Data Observatory
• Use CARTO’s spatially-enabled database for analysis
• Try it out without needing a CARTO account by using the Examples functionality

Common Uses

• Visualize spatial data programmatically as matplotlib images or embedded interactive maps
• Perform cloud-based spatial data processing using CARTO’s analysis tools
• Extract, transform, and Load (ETL) data using the Python ecosystem for getting data into and out of CARTO
• Data Services integrations using CARTO’s Data Observatory and other Data Services APIs

Try it out

The easiest way to try out cartoframes is to use the cartoframes example notebooks running in binder: https://mybinder.org/v2/gh/CartoDB/cartoframes/master?filepath=examples If you already have an API key, you can follow along and complete all of the example notebooks.

If you do not have an API key, you can use the Example Context to read the example data, make maps, and run arbitrary queries from the datasets there. The best place to get started is in the “Example Datasets” notebook found when running binder or downloading from the examples directory in the cartoframes GitHub repository.

Note

The example context only provides read access, so not all cartoframes features are available. For full access, Start a free 30 day trial or get free access with a GitHub Student Developer Pack.

More info

• Complete documentation: http://cartoframes.readthedocs.io/en/latest/
• Source code: https://github.com/CartoDB/cartoframes
• bug tracker / feature requests: https://github.com/CartoDB/cartoframes/issues

Note

cartoframes users must have a CARTO API key for most cartoframes functionality. For example, writing DataFrames to an account, reading from private tables, and visualizing data on maps all require an API key. CARTO provides API keys for education and nonprofit uses, among others. Request access at support@carto.com. API key access is also given through GitHub’s Student Developer Pack.

Install Instructions

To install cartoframes on your machine, do the following to install the latest version:

$ pip install cartoframes

cartoframes is continuously tested on Python versions 2.7, 3.5, and 3.6. It is recommended to use cartoframes in Jupyter Notebooks (pip install jupyter). See the example usage section below or notebooks in the examples directory for using cartoframes in that environment.

Virtual Environment

Using virtualenv

Make sure your virtualenv package is installed and up-to-date. See the official Python packaging page for more information.

To setup cartoframes and Jupyter in a virtual environment:

$ virtualenv venv
$ source venv/bin/activate
(venv) $ pip install cartoframes jupyter
(venv) $ jupyter notebook

Then create a new notebook and try the example code snippets below with tables that are in your CARTO account.

Using pipenv

Alternatively, pipenv provides an easy way to manage virtual environments. The steps below are:

1. Create a virtual environment with Python 3.4+ (recommended instead of Python 2.7)
2. Install cartoframes and Jupyter (optional) into the virtual environment
3. Enter the virtual environment
4. Launch a Jupyter notebook server

$ pipenv --three
$ pipenv install cartoframes jupyter
$ pipenv shell

Next, run a Python kernel by typing $ python, $ jupyter notebook, or however you typically run Python.

Native pip

If you install packages at a system level, you can install cartoframes with:

$ pip install cartoframes

Example usage

Data workflow

Get table from CARTO, make changes in pandas, sync updates with CARTO:

import cartoframes
# `base_url`s are of the form `https://{username}.carto.com/` for most users
cc = cartoframes.CartoContext(base_url='https://eschbacher.carto.com/',
                              api_key=APIKEY)

# read a table from your CARTO account to a DataFrame
df = cc.read('brooklyn_poverty_census_tracts')

# do fancy pandas operations (add/drop columns, change values, etc.)
df['poverty_per_pop'] = df['poverty_count'] / df['total_population']

# updates CARTO table with all changes from this session
cc.write(df, 'brooklyn_poverty_census_tracts', overwrite=True)

Write an existing pandas DataFrame to CARTO.

import pandas as pd
import cartoframes
df = pd.read_csv('acadia_biodiversity.csv')
cc = cartoframes.CartoContext(base_url=BASEURL,
                              api_key=APIKEY)
cc.write(df, 'acadia_biodiversity')

Map workflow

The following will embed a CARTO map in a Jupyter notebook, allowing for custom styling of the maps driven by TurboCARTO and CARTOColors. See the CARTOColors wiki for a full list of available color schemes.

from cartoframes import Layer, BaseMap, styling
cc = cartoframes.CartoContext(base_url=BASEURL,
                              api_key=APIKEY)
cc.map(layers=[BaseMap('light'),
               Layer('acadia_biodiversity',
                     color={'column': 'simpson_index',
                            'scheme': styling.tealRose(5)}),
               Layer('peregrine_falcon_nest_sites',
                     size='num_eggs',
                     color={'column': 'bird_id',
                            'scheme': styling.vivid(10)})],
       interactive=True)

Note

Legends are under active development. See https://github.com/CartoDB/cartoframes/pull/184 for more information. To try out that code, install cartoframes as:

pip install git+https://github.com/cartodb/cartoframes.git@add-legends-v1#egg=cartoframes

Data Observatory

Interact with CARTO’s Data Observatory:

import cartoframes
cc = cartoframes.CartoContext(BASEURL, APIKEY)

# total pop, high school diploma (normalized), median income, poverty status (normalized)
# See Data Observatory catalog for codes: https://cartodb.github.io/bigmetadata/index.html
data_obs_measures = [{'numer_id': 'us.census.acs.B01003001'},
                     {'numer_id': 'us.census.acs.B15003017',
                      'normalization': 'predenominated'},
                     {'numer_id': 'us.census.acs.B19013001'},
                     {'numer_id': 'us.census.acs.B17001002',
                      'normalization': 'predenominated'},]
df = cc.data('transactions', data_obs_measures)

CARTO Credential Management

Typical usage

The most common way to input credentials into cartoframes is through the CartoContext, as below. Replace {your_user_name} with your CARTO username and {your_api_key} with your API key, which you can find at https://{your_user_name}.carto.com/your_apps.

from cartoframes import CartoContext
cc = CartoContext(
    base_url='https://{your_user_name}.carto.com',
    api_key='{your_api_key}'
)

You can also set your credentials using the Credentials class:

from cartoframes import Credentials, CartoContext
cc = CartoContext(
    creds=Credentials(key='{your_api_key}', username='{your_user_name}')
)

Save/update credentials for later use

from cartoframes import Credentials, CartoContext
creds = Credentials(username='eschbacher', key='abcdefg')
creds.save()  # save credentials for later use (not dependent on Python session)

Once you save your credentials, you can get started in future sessions more quickly:

from cartoframes import CartoContext
cc = CartoContext()  # automatically loads credentials if previously saved

Experimental features

CARTOframes includes experimental features that we are testing for future releases into cartoframes core. These features exist as separate modules in vis. These features are stand-alone other than sometimes relying on some cartoframes utilities, etc. Vis features will also change often and without notice, so they should never be used in a production environment.

To import an experimental feature, like new vector maps, do the following:

from cartoframes.auth import Context
from cartoframes.viz import Map, Layer

context = Context()
Map(Layer('<table name>', '<carto vl style>', context=context))

• Overview
  • Features
  • Common Uses
  • Try it out
  • More info
  • Install Instructions
  • Example usage
• CartoContext
• Map Layers
• Layer Styling
• Example Functionalty
  • Data access functions
  • Example CartoContext
• Credentials
• Batch Jobs
• Cookbook
  • How to get census tracts or counties for a state
  • Get raw measures from the DO
  • Engineer your DO metadata if you already have GEOID or another geom_ref
  • How to get a matplotlib figure with four maps
  • Get a table as a GeoDataFrame
  • Skip SSL verification
  • Reading large tables or queries
  • Perform long running query if a time out occurs
  • Subdivide Data Observatory search region into sub-regions
• ETL with CARTOframes
  • Use cases
  • Examples
• Local Development Setup
  • Jupyter Notebook
  • CARTO VL
  • Airship
  • Run all the projects
  • Load JavaScript libraries locally
  • Reload changes

Indices and tables

• Index
• Module Index
• Search Page

Version

Version:0.10.1