Today, I checked all results possible in the basic model.
Summary:
I compare 4 conditions.
1. do not delete outliers, ANN model with just one layer
R2: 0.8564 0.8187 NRMSE: 0.0632 0.0672
2. do not delete outliers, ANN model with two layers
R2: 0.8611 0.8252 NRMSE: 0.0622 0.0660
3. delete outliers, ANN model with two layers
R2: 0.8494 0.8265 NRMSE: 0.0695 0.0689
4. delete outliers, ANN model with two layers, predict Vp and Vp/Vs together
R2: 0.8299 0.7964 NRMSE: 0.0712 0.0807
The second performs the best.
Tomorrow, I will continue to look for methods to improve the model such as deep ANN.
1. do not delete outliers, ANN model with just one layer
ReplyDeletewhat are you predicting?
also two sonic logs
Delete2. do not delete outliers, ANN model with two layers
ReplyDeletein this model ... what all inputs can be removed without affecting the quality of prediction?
we just have 13 inputs, I have not tried to delete them one by one to test.
Deletewhat if you predict one of them first followed by next..
ReplyDeleteI will try but since they are similar in accuracy when predicting together, I do not think that will help.
Delete