Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: Standard machine learning approaches outperform deep representation learning on phenotype prediction from transcriptomics data

Fig. 3

Performance by gene set and normalization. The performance of all models on each gene set (left column) and transform (right column). The results are divided by row into binary, multiclass, and survival tasks. For each gene set or normalization and task type, the gray points show the shifted statistics computed by subtracting the median of all models trained on the same task as a given model, and the black line is the median taken across all models and tasks

Back to article page