| from transformers import PretrainedConfig | |
| class ResnetConfig(PretrainedConfig): | |
| model_type = "custom_resnet50d" | |
| def __init__( | |
| self, | |
| block_type="bottleneck", | |
| layers: list[int] = [3, 4, 6, 3], | |
| num_classes: int = 1000, | |
| input_channels: int = 3, | |
| cardinality: int = 1, | |
| base_width: int = 64, | |
| stem_width: int = 64, | |
| stem_type: str = "", | |
| avg_down: bool = False, | |
| **kwargs, | |
| ): | |
| if block_type not in ["basic", "bottleneck"]: | |
| raise ValueError( | |
| f"block_type should be either 'basic' or 'bottleneck', but is {block_type}" | |
| ) | |
| if stem_type not in ["", "deep", "deep-tiered"]: | |
| raise ValueError( | |
| f"stem_type should be either '', 'deep' or 'deep-tiered', but is {stem_type}" | |
| ) | |
| self.block_type = block_type | |
| self.layers = layers | |
| self.num_classes = num_classes | |
| self.input_channels = input_channels | |
| self.cardinality = cardinality | |
| self.base_width = base_width | |
| self.stem_width = stem_width | |
| self.stem_type = stem_type | |
| self.avg_down = avg_down | |
| super().__init__(**kwargs) | |
| __all__ = [ | |
| "ResnetConfig", | |
| ] | |
| if __name__ == "__main__": | |
| resnet50d_config = ResnetConfig( | |
| block_type="bottleneck", stem_width=32, stem_type="deep", avg_down=True | |
| ) | |
| print(resnet50d_config) | |
| resnet50d_config.save_pretrained("resnet50d_config") | |