/ASI-revised

Attention-Based Spatial Interpolation for House Price Prediction

Primary LanguageJupyter Notebook

ASI

Attention-Based Spatial Interpolation for House Price Prediction

Requirements

  • python 3.8.10
  • tensorflow (>=2.5.0)

Data

  • Data: a numpy saved file (.npz) containing:
    • data['dist_eucli']
    • data['dist_geo']
    • data['idx_eucli']
    • data['idx_geo']
    • data['X_train']
    • data['X_test']
    • data['y_train']
    • data['y_test']

Installation

To install as a module:

$ conda create -n asi python=3.8.10
$ conda activate asi
$ git clone https://github.com/darniton/ASI
$ cd ASI
$ pip install -r requirements.txt
$ jupyter notebook

Replicating experiments

To replicate the experimental results:

  • ./notebooks: One notebook for each dataset