Skip to main content
. 2019 Jun 27;14(6):e0218951. doi: 10.1371/journal.pone.0218951

Fig 4. The impact of the top ranked features used on statistical model performance.

Fig 4

Training and testing datasets were created by splitting either datasets (A) or samples (B). Error bars were calculated by either mean±sd/√32 (A) or mean±sd/√80 (B).