This analysis cromprises structured and unstructured machine learning techniques on APD crime incidents data. We use various techniques including random forests, logistic regression, adaptive boosting, linear regression OLS, linear regression SGD, and MLP Regression.
- Classification and regression models: crime_class_regr.ipynb
- Clustering: crime_clust.ipynb
Python 3.*
Anaconda Python distribution (recommended)
Reference:
macOS: https://docs.continuum.io/anaconda/install/mac-os.html
Windows: https://docs.continuum.io/anaconda/install/windows
Reference: https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html
Example Code: conda env create -f crime-requirements.yml --name crime-env
Example Code:
macOS: conda activate crime-env
Windows: activate crime-env
Reference: https://jupyter-notebook.readthedocs.io/en/stable/
Example: jupyter notebook