Medical-SAM-Adapter

Medical-SAM-Adapter copied to clipboard

I would like to ask why this error occurred when I put it into my own data set. My data format labels are the same as yours. The only difference...

ChilsonHahe

Why didn't ResizeLongestSide be applied to the point prompt？

2

On line 122 of the function.py file.

xiawei20161308104

Point prompts have no work during training and inference, in other words, when inference, giving a point prompt at will can complete the segmentation

chaochenzhang

How do you train your model

1

SVT-Yang

The shape of the original input image is (256,256,600). After the previous processing, the shape of the current input x of ImageEncoderViT is (600,3,256,256). After self.patch_embed(), The shape of x...

jianjun0407

The adapter layer is not found in the part of the decoder

2

The adapter layer is not found in the part of the decoder, which does not match the description in the paper. Is this an oversight?

erliufashi

is it the same train process as isic dataset for amos dataset?

LinhuiHu

About Total score when evaluate

1

I found that during the test, the total score will be output. From an intuitive feeling, the larger the score, the better, but in your code logic, the tot should...

Bill-Ren

Suppose you use DiceMetric as criterion, do you compute Dice for every single batch and get average? How do you eliminate the affection of batch_size?

schmidtkk

Can you please let us know how you pre-trained the text encoder? Please provide the code for it.

sahilqure