MobileNet_V4

ConvNet Image Classification

class lucid.models.MobileNet_V4(cfg: Dict[str, dict], num_classes: int = 1000)

Overview

The MobileNet_V4 class provides the foundational architecture for the MobileNet-v4 model family. Building upon earlier MobileNet designs, it emphasizes both efficiency and flexibility, making it ideal for mobile and embedded applications. By utilizing a configurable design through a dictionary-based parameterization, this base class allows developers to easily experiment with different architectural variants, striking a balance between computational cost and model performance.

MobileNet-v4 architecture

Class Signature

class MobileNet_V4(nn.Module):
    def __init__(self, cfg: Dict[str, dict], num_classes: int = 1000) -> None

Parameters

  • cfg (Dict[str, dict]): A dictionary that holds the configuration for various network blocks. Each key in the dictionary represents a distinct module or layer group, and its corresponding value is another dictionary that specifies parameters (such as kernel size, expansion factors, filter counts, strides, etc.) for that module.

  • num_classes (int, optional): Specifies the number of output classes for the classification task. The default value is 1000, which is typically used for datasets like ImageNet.