Figure 10 | Scientific Reports

Figure 10

From: Evaluation of two semi-supervised learning methods and their combination for automatic classification of bone marrow cells

Figure 10

Original architecture for this study. The architecture consisted of total of eight layers of two dimensional convolution. The figures under each convolution layer indicate the size (height and width) and number of future maps and layers. “Conv2D” refers to 2-dimentional convolution. “BatchNorm” refers to batch normalization. “SE block” refers to Squeeze-and-Excitation block.

Back to article page