get_efficientnet_kwargs¶

get_efficientnet_kwargs(channel_multiplier=1.0, depth_multiplier=1.0, drop_rate=0.2)[source]¶

Generate configuration parameters for EfficientNet.

Parameters:

  • channel_multiplier (float) – Multiplier for number of channels per layer. Defaults to 1.0.

  • depth_multiplier (float) – Multiplier for number of repeats per stage. Defaults to 1.0.

  • drop_rate (float) – Dropout rate. Defaults to 0.2.

Return type:

dict[str, Any]

Reference implementation:

https://github.com/tensorflow/tpu/blob/master/models/official/efficientnet/efficientnet_model.py

Paper:

https://arxiv.org/abs/1905.11946

EfficientNet parameters: - ‘efficientnet-b0’: (1.0, 1.0, 224, 0.2) - ‘efficientnet-b1’: (1.0, 1.1, 240, 0.2) - ‘efficientnet-b2’: (1.1, 1.2, 260, 0.3) - ‘efficientnet-b3’: (1.2, 1.4, 300, 0.3) - ‘efficientnet-b4’: (1.4, 1.8, 380, 0.4) - ‘efficientnet-b5’: (1.6, 2.2, 456, 0.4) - ‘efficientnet-b6’: (1.8, 2.6, 528, 0.5) - ‘efficientnet-b7’: (2.0, 3.1, 600, 0.5) - ‘efficientnet-b8’: (2.2, 3.6, 672, 0.5) - ‘efficientnet-l2’: (4.3, 5.3, 800, 0.5)

Parameters:

  • channel_multiplier (float) – Multiplier to number of channels per layer. Defaults to 1.0.

  • depth_multiplier (float) – Multiplier to number of repeats per stage. Defaults to 1.0.

  • drop_rate (float) – Dropout rate. Defaults to 0.2.

Returns:

Dictionary containing EfficientNet configuration parameters

Return type:

dict[str, Any]

Example

>>> kwargs = get_efficientnet_kwargs(
...  channel_multiplier=1.2,
...  depth_multiplier=1.4,
... )