#!/usr/bin/env python 
# -*- coding: utf-8 -*-
"""
@project: 
@File    : temple_part
@Author  : qiqq
@create_time    : 2023/5/22 22:38
"""

import torch
from torch import nn
import torch.nn.functional as F
import numpy as np
import math

# class upmodel(nn.Module):
#
#     def __init__(self,
#                  in_channels,
#                  out_channels,
#                  in_feat_output_strides=(4, 8, 16, 32),
#                  out_feat_output_stride=4,
#                  ):
#         super(upmodel, self).__init__()
#         self.blocks = nn.ModuleList()
#         for in_feat_os in in_feat_output_strides:
#             num_upsample = int(math.log2(int(in_feat_os))) - int(math.log2(int(out_feat_output_stride)))
#
#             num_layers = num_upsample if num_upsample != 0 else 1
#
#             self.blocks.append(nn.Sequential(*[
#                 nn.Sequential(
#                     nn.Conv2d(in_channels if idx == 0 else out_channels, out_channels, 3, 1, 1, bias=False),
#                     nn.BatchNorm2d(out_channels),
#                     nn.ReLU(inplace=True),
#                     nn.UpsamplingBilinear2d(scale_factor=2) if num_upsample != 0 else nn.Identity(),
#                 )
#                 for idx in range(num_layers)]))
#
#     def forward(self, feat_list: list):
#         inner_feat_list = []
#         for idx, block in enumerate(self.blocks):
#             decoder_feat = block(feat_list[idx])
#             inner_feat_list.append(decoder_feat)
#
#         out_feat = sum(inner_feat_list) / 4.  #待定
#         return out_feat
#
#
# impo1=torch.rand(1,64,128,128)
# impo2=torch.rand(1,256,64,64)
# impo1=torch.rand(1,512,32,32)
# impo1=torch.rand(1,768,16,16)
#
# lis=[torch.rand(1,64,128,128),
# torch.rand(1,256,64,64),
# torch.rand(1,512,32,32),
# torch.rand(1,768,16,16)
#      ]
#
# model=upmodel()