Skip to main content

Table 1 The training environment and training strategy

From: Development of a modified 3D region proposal network for lung nodule detection in computed tomography scans: a secondary analysis of lung nodule datasets

Training environment

OS: Windows 10 (64-bit)

CPU: Intel Core i7-9700 H @ 3.00 GHz

GPU: Single NVIDIA GeForce RTX 2070–8GB

RAM: 24GB

Deep learning framework: Pytorch

Training strategy

Optimizer: Ranger (with Gradient Centralization)

Learning rate: 0.001

Epoch: 200

Data augmentation method:

Randomly fliping in xy plane

Randomly rotating 0 ~ 180 degrees on the z-axis

Random scaling 0.75 ~ 1.25 times