Analysis of Moscow rental property's pricing

The project presents building a model that predicts rental price per square meter for 1 to 5-room flats in Moscow. The data for training were scraped from advertisements on a popular classified avito.ru. Scraping was conducted daily from August 1 to August 14, leading to 14 thousand observations. Random forest is used for prediction and achieves mean absolute percentage error of 16%.

Description of the files (in the order of execution):

1. scraping.py - scraping data from Avito and saving them to pickles
2. stations.py - downloading information about Moscow underground stations and computing their distances to the city center
3. make_dataset.py - constructing a dataframe for exploratory data analysis from pickle files
4. eda.ipynb - exploratory analysis of the prices, publications flow and commissions
5. modelling.ipynb - hyperparameter optimization with cross-validation
6. build_features.py - constructing a dataframe for training from pickle files
7. train_model.py - training random forest with hyperparameters found in modelling.py
8. predict_model.py - making predictions with the trained model

Only eda.ipynb and modelling.ipynb contain comments.

├── README.md          <- The top-level README for developers using this project.
├── data
│   ├── interim        <- Intermediate data that has been transformed.
│   ├── processed      <- The final, canonical data sets for modeling.
│   └── raw            <- The original, immutable data dump.
│
├── models             <- Trained and serialized models
│
├── notebooks          <- Jupyter notebooks.
│   |
|   ├── eda.ipynb      <- Exploratory data analysis
|   └── modelling.ipynb<- Hyperparameter optimization
|
├── requirements.txt   <- The requirements file for reproducing the analysis environment
├── src                <- Source code for use in this project.
    ├── __init__.py    <- Makes src a Python module
    │
    ├── data           <- Scripts to download or generate data
    │   ├── make_dataset.py
    |   ├── scraping.py
    |   └── stations.py
    │
    ├── features       <- Scripts to turn raw data into features for modeling
    │   ├── build_features.py
    │   
    |   
    ├── models         <- Scripts to train models and then use trained models to make
        │                 predictions
        ├── predict_model.py
        └── train_model.py

Project based on the cookiecutter data science project template. #cookiecutterdatascience