This is the official documentation for Quandl's Python Package. The package can be used to interact with the latest version of the Quandl RESTful API. This package is compatible with python v2.7.x and v3.x+.
Please see this readme for more information and upgrade instructions: 2.x series transition notes
The installation process varies depending on your python version and system used. However in most cases the following should work:
pip install quandl
Alternatively on some systems python3 may use a different pip executable and may need to be installed via an alternate pip command. For example:
pip3 install quandl
|api_key||Your access key||
|use_retries||Whether API calls which return statuses in
|number_of_retries||Maximum number of retries that should be attempted. Only used if
|max_wait_between_retries||Maximum amount of time in seconds that should be waited before attempting a retry. Only used if
|retry_backoff_factor||Determines the amount of time in seconds that should be waited before attempting another retry. Note that this factor is exponential so a
|retry_status_codes||A list of HTTP status codes which will trigger a retry to occur. Only used if
||[429, 500, 501, 502, 503, 504, 505, 506, 507, 508, 509, 510, 511]|
import quandl quandl.ApiConfig.api_key = 'tEsTkEy123456789'
By default, SSL verification is enabled. To bypass SSL verification
quandl.ApiConfig.verify_ssl = False
Save local key to
import quandl quandl.save_key("supersecret") print(quandl.ApiConfig.api_key)
Load the API Key without exposing the key in the script or notebook
import quandl quandl.read_key() print(quandl.ApiConfig.api_key)
Set a custom location for the API key file, e.g. store the externally outside a docker container
import quandl quandl.save_key("ourcorporateapikey", filename="/srv/data/somecontainer/.corporatequandlapikey")
and call within the docker container with mount point at
import quandl quandl.read_key(filepath="/data/.corporatequandlapikey")
There are two methods for retrieving data in Python: the Quick method and the Detailed method. The latter is more suitable to application programming. Both methods work with Quandl's two types of data structures: time-series (dataset) data and non-time series (datatable).
The following quick call can be used to retrieve a dataset:
import quandl data = quandl.get('NSE/OIL')
This example finds all data points for the dataset
NSE/OIL and stores them in a pandas dataframe. You can then view the dataframe with data.head().
A similar quick call can be used to retrieve a datatable:
import quandl data = quandl.get_table('ZACKS/FC', ticker='AAPL')
This example retrieves all rows for
ticker='AAPL' and stores them in a pandas dataframe. Similarly you can then view the dataframe with data.head().
Note that in both examples if an
api_key has not been set you may receive limited or sample data. You can find more details on these quick calls and others in our Quick Method Guide.
Currently, Quandl debug logging is limited in scope. However, to enable debug logs you can use the following snippet.
import quandl import logging logging.basicConfig() # logging.getLogger().setLevel(logging.DEBUG) # optionally set level for everything. Useful to see dependency debug info as well. quandl_log = logging.getLogger("quandl") quandl_log.setLevel(logging.DEBUG)
Our API can provide more than just data. It can also be used to search and provide metadata or to programmatically retrieve data. For these more advanced techniques please follow our Detailed Method Guide.
If you wish to work on local development please clone/fork the git repo and use
pip install -r requirements.txt to setup the project.
We recommend the following tools for testing any changes:
The following are instructions for running our tests:
pip install tox virtualenv
python setup.py install
Once you have all required packages installed, you can run tests locally with:
Running all tests locally
python -W always setup.py -q test
Running an individual test
python -m unittest test.[test file name].[class name].[individual test name]`
python -m unittest -v test.test_datatable.ExportDataTableTest.test_download_get_file_info
We would suggest downloading the data in raw format in the highest frequency possible and performing any data manipulation in pandas itself.
See this link for more information about timeseries in pandas.
To release the package, you can follow the instructions on this page