keras-io/3D_CNN_Pneumonia

Model description

This model is a 3D convolutional neural network model trained to predict the presence of viral pneumonia in computer tomography scans.

Spaces link

Keras Example Link

Training and evaluation data

Subset of the MosMedData: Chest CT Scans with COVID-19 Related Findings

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• optimizer: {'name': 'Adam', 'learning_rate': {'class_name': 'ExponentialDecay', 'config': {'initial_learning_rate': 0.0001, 'decay_steps': 100000, 'decay_rate': 0.96, 'staircase': True, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
• training_precision: float32

Model Plot

View Model Plot