Installation#

Dependencies#

All Python Version Minimal Dependencies#

Required Dependencies:

Python (>= 3.9)
pandas (>= 1.5.1)

Optional Dependencies:

dtoolkit.transformer requires dependencies
- Scikit-learn (>= 1.4)
- packaging
dtoolkit.geoaccessor requires dependencies
- GeoPandas (>= 0.12.1)

Different Python Version Minimal Dependencies#

Dtoolkit support Python version from 3.9 to 3.10. Therefore each version supports minimal dependencies is different. You could check the dependencies list from following.

DToolKit Requires Python Version History#

DToolKit 0.0.1 requires Python 3.7 or 3.8.
DToolKit 0.0.2 to 0.0.5 require Python 3.7 to 3.9.
DToolKit 0.0.6 to 0.0.16 require Python 3.7 to 3.10.
DToolKit 0.0.17 and 0.0.18 require Python 3.8 to 3.10.
DToolKit 0.0.19 and later require Python 3.8 or 3.11.
DToolKit 0.0.21 and later require Python 3.9 or newer.

Install with Mamba#

To install all DToolKit’s dependencies, we recommend to use Mamba: The Fast Cross-Platform Package Manager. The advantage of using the mamba is that it provides pre-built binaries for all the required and optional dependencies of DToolKit for all platforms and faster to download them.

Mamba will help you get out of many troubles such as the versions of dependencies, conflicts from channels or environment itself, downloading from where or speed, and so on.

You always can delete that problem environment and swift to another good environment.

The following commands create a new environment with the name dtoolkit_env, then installs dependencies and DToolKit in it:

From Command#

$ mamba create -n dtoolkit_env
$ mamba activate dtoolkit_env
$ mamba install python=3 pandas scikit-learn packaging geopandas
(dtoolkit_env)$ pip install my-data-toolkit

From YAML#

Save the following environment.yaml YAML file to local.

name: dtoolkit
channels:
  - conda-forge
dependencies:
  # required
  - pandas >= 1.5.1
  # optional
  - scikit-learn >= 1.2
  - packaging
  - geopandas >= 0.12.1
  - geopy
  - jenkspy
  - topojson
  - h3-py
  - joblib
  - rapidfuzz

  - pip
  - pip:
      - my-data-toolkit

Create the environment from YAML.

$ mamba env create -f environment.yaml

Install from PyPI#

$ pip install my-data-toolkit

Install from Source#

Warning

This’s a latest version but not a stable version.

$ pip install git+https://github.com/Zeroto521/my-data-toolkit