Using autogluon to forecast house price

AutoGluon Compatibility and scalability of , It is easy to carry out scientific research algorithms into large-scale experiments .

Three application fields

• image（image classification、object detection）
• text（text classification）
• tabular data（tabular prediction）

Two functions

• Automatic parameter adjustment
  • Support PyTorch
  • Supported search strategies include random search、grid search、RL、Bayesian optimization etc.
• NAS（ Support only image classification, At present, only ENAS） 

from autogluon.tabular import TabularDataset,TabularPredictor, predictor
from catboost import train
import numpy as np
# # Training 
train_data = TabularDataset('/cloud_disk/users/huh/projects/csdn_send_data/autogluon/house_price_kaggle/train.csv')
id, label = 'Id','SalePrice'
#  Data preprocessing 
large_val_cols = ['LotFrontage','LotArea','OverallQual','OverallCond','YearBuilt','YearRemodAdd','WoodDeckSF','OpenPorchSF','EnclosedPorch','3SsnPorch','MiscVal','MoSold','YrSold']


for c in large_val_cols +[label]:
    train_data[c] = np.log(train_data[c]+1)
# autogluon Feature extraction can be done , However, some manual pretreatment should be properly added 
#  Use multimodel This option to use transformer To extract features , And do multi model fusion 
#  Then make a multi-layer model ensemble To get better accuracy 
predictor = TabularPredictor(label=label).fit(train_data.drop(columns=[id]))#  nothing multimodel

Dataset Download ：

link : https://pan.baidu.com/s/1XsKH3k4DbZ72MpU3G9WNDQ password : 1nf7