I2U-Net

I2U-Net copied to clipboard

3D网络模型

1

您好，我想问一下，可以提供3D的网络模型吗

z1h2a3o

Training problem

11

Dear Author: I use your model for polyp segmentation,but during the training, the loss don't change, which makes me confused.

lsh-med

 如图，在金字塔部分，空洞卷积的dilate4_out，dilate4_out为什么要串行依次执行不同空洞率的空洞卷积呢？dilate 1，2，3不是已经执行过了吗？为什么dilate4不单独执行4对应的空洞率的卷积，而是将从空洞率1，2，3的卷积都执行一遍再传输到dilate4的卷积？

085410

loss value is negative

2

The ISIC Dice score: 0.8744; The ISIC JC score: 0.7953 [0, 1.4520071135297217, 1.61416205042743, 1.6508494510681295, 1.6697128879217] the best model will be saved at ./saved_models_0920/ISIC2018\folder3\I2U_Net_L/best_score_ISIC2018_checkpoint.pth.tar 0%| | 0/227 [00:00

Kingxudong

test setup failed file E:\BTBU人工智能\研1\视听文组项目\2024.10.18医学图像分割模型\I2U-Net-main\train_and_test_isic.py, line 180 def test_isic(test_loader, model, args, test_acc_log, date_type, save_img=True): E fixture 'test_loader' not found > available fixtures: anyio_backend, anyio_backend_name, anyio_backend_options, cache, capfd, capfdbinary, caplog, capsys,...

sadadsaada

Can you provide Polyp datasets segmentation results as I am trying to reproduce the results given in the paper. However, the results I get are far from the results given...

juingzhou

作者您好，可以请您提供一下您论文中表2中，CVC-ClinicDB和Kvasir的数据集预处理方法和训练集测试集划分病例txt吗？我想引用您论文里面这张表的数据

w573352

我用下面的脚本在CVC-CLINICDB上做测试，但是最后dice只有0.85左右。 ```python from Models.I2U_Net import I2U_Net_L from tqdm import tqdm import torch from torch.optim import lr_scheduler from utils.dice_loss import get_soft_label from utils.dice_loss_github import SoftDiceLoss_git, CrossentropyND from Dataset.Dataset import Dataset from...

Aurora2580