how to compute bilinear interpolate (eg: segment anything 2 include bilinear interpolate)

[yysu-888]

import torch.nn.functional as F
F.interpolate(...,mode = "bilinear")

[YavorGIvanov]

Previously I had to just preprocess the image independent of ggml - https://github.com/YavorGIvanov/sam.cpp/blob/81002818eb0e2cb3b9a523286b067f80f8424431/sam.cpp#L355

However, I am not sure whether there is easier way to do it in ggml

[yysu-888]

Previously I had to just preprocess the image independent of ggml - https://github.com/YavorGIvanov/sam.cpp/blob/81002818eb0e2cb3b9a523286b067f80f8424431/sam.cpp#L355

However, I am not sure whether there is easier way to do it in ggml

I mean sam2 not sam,https://github.com/facebookresearch/sam2/blob/main/sam2/modeling/backbones/image_encoder.py, in line 116 F.interpolate (fpn_interp_model: str = "bilinear")as follow
top_down_features = F.interpolate(
prev_features.to(dtype=torch.float32),
scale_factor=2.0,
mode=self.fpn_interp_model,
align_corners=(
None if self.fpn_interp_model == "nearest" else False
),