I have two datasets A and B which are exactly the same in terms of the number of columns, name of columns and the values. The only difference is the order of those columns. I then train the LightGBM model on each of the two datasets with the following steps
- Divide each dataset into training and testing (use the same random seed and ratio for both A and B)
- Leave the hyperparameters as pretty much default
- Set a random state as a fixed number (for reproduction)
- Tune the learning_rate using a Grid Search
- Train a LightGBM model on the training set and test it on the testing set
- Learning rate with the best performance on the testing set will be chosen
The output models on the two datasets are very different, which makes me thinks that the order of columns does affect the performance of the model training using LightGBM.
Do you know why this is the case?