Skip to content

weichi21/PyTorch-Neural-network-median-house-price-prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

PyTorch Median House Price Prediction

  • Objective: Leveraging deep learning architecture to construct a prediction model for median house prices.

Dataset can be downloaded/loaded using the func:sklearn.datasets.fetch_california_housing function.

Here I implemented:

  1. Data Standardization
  2. Create simple linear regression neural network
  3. Create dynamic non-linear regression neural network for hyperparameter tuning
  4. Comparing the convergence performance of different loss functions
  5. Hyperparameter tuning and cross-validation of hidden layers, number of neurons, activation function, and learning rate

Conclusion

  • Best parameters: {'activation function': Tanh(), 'number of neurons': 16, 'hidden layers': 2, 'learning rate': 0.05}
  • Best mean CV score (MAPE): 20.5663
  • Test score (MAPE): 20.1813
  • Test R2 Score: 0.792
  • Average Accuracy: 79.82%