fMRI-foundation-model/src/mae_utils/pos_embed.py

# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.

# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
# --------------------------------------------------------
# Position embedding utils
# --------------------------------------------------------

import util.logging as logging
import numpy as np
import torch


logger = logging.get_logger(__name__)


# --------------------------------------------------------
# Interpolate position embeddings for high-resolution
# References:
# DeiT: https://github.com/facebookresearch/deit
# --------------------------------------------------------
def interpolate_pos_embed(model, checkpoint_model):
    if "pos_embed" in checkpoint_model:
        pos_embed_checkpoint = checkpoint_model["pos_embed"]
        embedding_size = pos_embed_checkpoint.shape[-1]
        num_patches = model.patch_embed.num_patches
        num_extra_tokens = model.pos_embed.shape[-2] - num_patches
        # height (== width) for the checkpoint position embedding
        orig_size = int((pos_embed_checkpoint.shape[-2] - num_extra_tokens) ** 0.5)
        # height (== width) for the new position embedding
        new_size = int(num_patches**0.5)
        # class_token and dist_token are kept unchanged
        if orig_size != new_size:
            print(
                "Position interpolate from %dx%d to %dx%d"
                % (orig_size, orig_size, new_size, new_size)
            )
            extra_tokens = pos_embed_checkpoint[:, :num_extra_tokens]
            # only the position tokens are interpolated
            pos_tokens = pos_embed_checkpoint[:, num_extra_tokens:]
            pos_tokens = pos_tokens.reshape(
                -1, orig_size, orig_size, embedding_size
            ).permute(0, 3, 1, 2)
            pos_tokens = torch.nn.functional.interpolate(
                pos_tokens,
                size=(new_size, new_size),
                mode="bicubic",
                align_corners=False,
            )
            pos_tokens = pos_tokens.permute(0, 2, 3, 1).flatten(1, 2)
            new_pos_embed = torch.cat((extra_tokens, pos_tokens), dim=1)
            checkpoint_model["pos_embed"] = new_pos_embed