feature extraction stages

[jawi289o]

I have a question about understanding the structure and functionality of backbone stages. I want to know how to identify which layers belong to each stage.
For example, MobileNetV3 has 5 feature extraction stages (or downsampling stages), and I’d like to understand what layers or operations are included in each.
Stage 1: What kind of layers (e.g., convolutions, activations) are included?
Stage 2, Stage 3, etc.: What layers or operations are present here?
I want to better understand the breakdown of these stages to analyze how features are processed and extracted.

[james77777778]

Hi @jawi289o
Thanks for asking! You can check the variable features in all models to identify which tensor belongs to the extraction stage.

For example:

• Initialize features: https://github.com/james77777778/keras-image-models/blob/main/kimm/_src/models/mobilenet_v3.py#L141-L142
• "STEM": https://github.com/james77777778/keras-image-models/blob/main/kimm/_src/models/mobilenet_v3.py#L158
• "BLOCK": https://github.com/james77777778/keras-image-models/blob/main/kimm/_src/models/mobilenet_v3.py#L229