Table 5 Ten-fold cross-validation of livestock densities estimation in different regions (N represents the number of training samples (counties)).
From: High-resolution livestock seasonal distribution data on the Qinghai-Tibet Plateau in 2020
Region | Response factor | Pasture mask | N | R2 | MAE | MSE |
---|---|---|---|---|---|---|
QTP | Ln(Livestock density) | All pastures | 164 | 0.700 | 0.510 | 0.250 |
Ln(Cattle density) | All pastures | 164 | 0.706 | 0.612 | 0.565 | |
Ln(Sheep density) | All pastures | 164 | 0.742 | 0.783 | 0.515 | |
Tibet | Ln(Livestock density) | All pastures | 74 | 0.684 | 0.385 | 0.249 |
Ln(Cattle density) | All pastures | 74 | 0.809 | 0.401 | 0.274 | |
Ln(Sheep density) | All pastures | 74 | 0.712 | 0.646 | 0.899 | |
Qinghai | Ln(Livestock density) | All pastures | 44 | 0.709 | 0.448 | 0.447 |
Ln(Cattle density) | All pastures | 44 | 0.678 | 0.635 | 0.685 | |
Ln(Sheep density) | All pastures | 44 | 0.830 | 0.519 | 0.467 |