Research Article

Failure Analysis of Static Analysis Software Module Based on Big Data Tendency Prediction

Table 3

Hyperparameter settings of gcForest and DSF.

Deep Forest (gcForest)Deep stacked forest (DSF)

Forest type:Forest type:
Random forest, completely random tree forestRandom forest, completely random tree forest
Forest during multigranularity scanning:Forest at random sampling:
Number of forests: 2Number of forests: 2
Trees in the forest: 20Trees in the forest: 20
The size of the sliding window:The size of the sliding window:
Number of forests: 4Random sampling times: 200, 100, 50 cascade forest:
Trees per tree in the forest: 200Number of forests: 4, trees per tree in the forest: 200