justchenhao / BIT_CD

Official Pytorch Implementation of "Remote Sensing Image Change Detection with Transformers"

justchenhao/BIT_CD Issues

DSIFN dataset
Updated 2 months ago5
训练报错，提示找不到文件
Updated 2 months ago2
請問如何detect?
Updated 2 months ago1
ValueError: Input and output must have the same number of spatial dimensions, but got input with with spatial dimensions of [256, 256] and output size of torch.Size([1, 256, 256, 3]). Please provide input tensor in (N, C, d1, d2, ...,dK) format and output size in (o1, o2, ...,oK) format.
Updated 3 months ago7
Some problems about the image size of training process
Closed a year ago1
F1 values of WHU-CD and DSIFN-CD
Updated 6 months ago9
出现Windows fatal exception: code 0xc0000139，test setup failed
Updated 6 months ago3
WHU-CD dataset
Updated 7 months ago11
Something for ‘update the best model’
Updated 7 months ago
Is the IOU_1 value higher than the paper？
Updated 8 months ago4
Something for 'stride' of the Resnet18 in Backbone
Closed 8 months ago1
TypeError: cannot pickle '_io.TextIOWrapper' object
Updated 8 months ago
demo的问题
Updated 8 months ago
python demo.py ModuleNotFoundError: No module named 'torchvision.models.utils'
Updated 10 months ago2
detect my own image
Closed a year ago
关于可视化问题
Updated a year ago6
How could I upgrade the accuracy?
Updated a year ago7
关于训练过程F1的疑问
Updated a year ago
The number of parameters and FLOPs!
Updated a year ago2
How to preprocess the WHU-CD dataset
Closed a year ago
index
Updated a year ago
I have a question.
Updated a year ago1
How you implemented the ablation experiment
Updated a year ago
Training Error
Closed a year ago3
size mismatch for transformer_decoder.layers.0.0.fn.fn.to_q.weight: copying a param with shape torch.Size([512, 32]) from checkpoint, the shape in current model is torch.Size([64, 32]).
Updated a year ago12
Nan loss. Sometimes weight initialize to -Inf when training from scratch
Closed a year ago
Issue about Unsupervised Learning
Updated a year ago
feature_map
Updated a year ago1
data
Updated 2 years ago15
The data in the paper is obtained with base_transformer_pos_s4_dd8 or base_transformer_pos_s4_dd8_dedim8 in the run_cd.sh file
Updated 2 years ago1
检测结果全黑
Updated 2 years ago1
WHU-CD
Updated 2 years ago5
ValueError: attempted relative import beyond top-level package
Updated 2 years ago
train error
Updated 2 years ago
Lower index
Closed 2 years ago2
IndexError: boolean index did not match indexed array along dimension 0
Closed 2 years ago1
Params. and FLOPs
Closed 2 years ago
How to download WHU-CD dataset?
Closed 2 years ago1
F1 Metric
Updated 2 years ago7
Transformer pre-training
Updated 3 years ago
请问对于变化检测到底是基于语义方式例如还是度量方式（STANet）好点呢？
Updated 3 years ago
关于测试数据输出可视化的问题
Updated 3 years ago
cannot reproduce to this paper accuracy
Updated 3 years ago4
测试指标比论文高
Closed 3 years ago2
backbone
Closed 3 years ago1
About the eval results
Closed 3 years ago1
Data
Updated 3 years ago
Fine tuning with data of lower resolution
Updated 3 years ago
test
Closed 3 years ago