def __init__()

in mask_former/modeling/backbone/swin.py [0:0]

• 48 lines of code
• 1 McCabe index (conditional complexity)

    def __init__(self, cfg, input_shape):

        pretrain_img_size = cfg.MODEL.SWIN.PRETRAIN_IMG_SIZE
        patch_size = cfg.MODEL.SWIN.PATCH_SIZE
        in_chans = 3
        embed_dim = cfg.MODEL.SWIN.EMBED_DIM
        depths = cfg.MODEL.SWIN.DEPTHS
        num_heads = cfg.MODEL.SWIN.NUM_HEADS
        window_size = cfg.MODEL.SWIN.WINDOW_SIZE
        mlp_ratio = cfg.MODEL.SWIN.MLP_RATIO
        qkv_bias = cfg.MODEL.SWIN.QKV_BIAS
        qk_scale = cfg.MODEL.SWIN.QK_SCALE
        drop_rate = cfg.MODEL.SWIN.DROP_RATE
        attn_drop_rate = cfg.MODEL.SWIN.ATTN_DROP_RATE
        drop_path_rate = cfg.MODEL.SWIN.DROP_PATH_RATE
        norm_layer = nn.LayerNorm
        ape = cfg.MODEL.SWIN.APE
        patch_norm = cfg.MODEL.SWIN.PATCH_NORM

        super().__init__(
            pretrain_img_size,
            patch_size,
            in_chans,
            embed_dim,
            depths,
            num_heads,
            window_size,
            mlp_ratio,
            qkv_bias,
            qk_scale,
            drop_rate,
            attn_drop_rate,
            drop_path_rate,
            norm_layer,
            ape,
            patch_norm,
        )

        self._out_features = cfg.MODEL.SWIN.OUT_FEATURES

        self._out_feature_strides = {
            "res2": 4,
            "res3": 8,
            "res4": 16,
            "res5": 32,
        }
        self._out_feature_channels = {
            "res2": self.num_features[0],
            "res3": self.num_features[1],
            "res4": self.num_features[2],
            "res5": self.num_features[3],
        }