zfnet¶
The zfnet function in lucid.models provides a convenient way to create an instance of the ZFNet module, a convolutional neural network designed for image classification with enhanced feature learning.
Total Parameters: 62,357,608
Function Signature¶
@register_model
def zfnet(num_classes: int = 1000, **kwargs) -> ZFNet
Parameters¶
num_classes (int, optional): The number of output classes for classification. Default is 1000.
kwargs (dict, optional): Additional keyword arguments to customize the ZFNet module.
Returns¶
ZFNet: An instance of the ZFNet module configured with the specified number of classes and any additional arguments.
Examples¶
Creating a Default ZFNet Model
import lucid.models as models
# Create a ZFNet model with 1000 output classes
model = models.zfnet()
print(model) # Displays the ZFNet architecture
Custom Number of Classes
# Create a ZFNet model with 10 output classes
model = models.zfnet(num_classes=10)
print(model) # Displays the ZFNet architecture with modified output