To fit the discriminative model, we transform the original space into a bank of DoG images (Marr and Hildreth 1980), where db is the bth filter bank image, b ε H | b = α 0 + α 1 r, r = 1, 2, …, n, α 0 is the starting bank, α 1 is the spacing of the banks, and n is the number of banks.
Based on this model, we transformed the fitted log-transformed odds to a score ranging from 0 to 200 using the intercept a = 190.7 and linear coefficient b = 34.0.
To facilitate parameter estimation of the model, we transformed ψ cos(2 πωt j − θ) into a linear form [ 19].
In order to find the best-fit model, we transformed the continuous variables into binary variables using the Youden index-J of the receiver operating characteristic (ROC) curve to maximize their discriminative powers of the best-fit model (e.g., age ≥67 years, BMI <22 kg/m).
Based on these realistic models, we transform the link scheduling problem to a vertex-coloring problem, using the concept of conflict graph which profile interferences in a co-channel wireless network.
To correlate the human data with the output of the saliency models, we transform the eye-fixation data to fixation- distance maps (see Fig. 5).
In the linear regression models we transformed dimethoate levels using the square root transformation to improve the normality of its distribution.
Starting from a conceptual UML model, we first transform the model before refining it into executable code.
"In the model we basically transform Earth's atmosphere into little 'cubes' and in each cube the fundamental equations controlling motion, energy and continuity of the atmosphere are solved," said GMAO meteorologist Oreste Reale in NASA's post.
With the constructed model, we can transform a raw DIC microscope image into a nuclear-enhanced probability map, which achieves very high contrast between the nuclear and cytoplasmic regions.
Instead of directly discovering a structured process model, we first apply a well-known heuristic that discovers accurate but oftentimes unstructured (and even unsound) process models, and then we transform the resulting process model into a structured (and sound) one.
