Skip to main content
. 2019 Jul 10;20(14):3389. doi: 10.3390/ijms20143389

Figure 2.

Figure 2

Absolute (a) and percentage (b) of R2 increase between ST-DNN and MT-DNN and the Cubist benchmark for the solubility endpoint. DNN, deep neural network; HCl, hydrochloric acid; MT, multi-task; PBS, phosphate-buffered saline; SIF, simulated intestinal fluid; ST, single task.