OpenGVLab/SAM-Med2D

Issues

• How to train with multiple mask labels

  #64 opened 4 months ago by ncp-nice
  4

• About the dataset size

  #71 opened 3 months ago by a-green-hand-jack
  0

• Data Unzip Problem

  #70 opened 3 months ago by zihui-debug
  0

• the problemdata

  #54 opened 3 months ago by liubaoning111
  1

• data_demo

  #24 opened 3 months ago by apuomline
  1

• 公开的模型权重类型

  #35 opened 3 months ago by wade0604
  1

• Request for requirements.txt

  #45 opened 3 months ago by ariharasudhanm
  1

• batchsize的大小

  #53 opened 3 months ago by 21-10-4
  2

• why take 3 channels png image as input?

  #55 opened 3 months ago by skywalker0523
  1

• Output the mask

  #59 opened 3 months ago by ChaoyiDu
  1

• the condition of sigmoid in metrics.py

  #60 opened 3 months ago by deep-geo
  1

• 推理问题

  #66 opened 3 months ago by lpc-97
  1

• Exporting the operator repeat_interleave to ONNX opset version 12 is not supported.

  #69 opened 3 months ago by dw1360585641
  0

• how to prediction without using main. py of SAM-Med2D onnxruntime？how to modify test.py to predicte to get mask？

  #68 opened 3 months ago by dw1360585641
  0

• adapter layer

  #22 opened a year ago by 98zhenyu
  2

• Validation Datasets and Splits

  #65 opened 4 months ago by timat33
  0

• Question regarding the dataset at HuggingFace

  #63 opened 4 months ago by anwai98
  1

• Some issue about training proecss

  #62 opened 4 months ago by Hhankyangg
  0

• Efficiency of the adapter

  #61 opened 4 months ago by ariharasudhanm
  1

• How to generate 2D slices from 3D CT or MRI

  #57 opened 5 months ago by lzl2040
  1

• The later code will overwrite the previous code and cause the adapter to freeze.

  #58 opened 5 months ago by ncp-nice
  1

• Partitioned datasets

  #51 opened 4 months ago by Rey-nard
  2

• Training without prompting

  #28 opened a year ago by ariharasudhanm
  5

• train on own datasets

  #26 opened 6 months ago by Genniew
  2

• data_demo

  #23 opened 6 months ago by apuomline
  1

• The preprocess code for converting .nii.gz to .png？

  #19 opened a year ago by JWZhao-uestc
  1

• Multiple masks abnormality

  #46 opened 9 months ago by sahil02235
  1

• Regarding testing issues

  #56 opened 6 months ago by Tzx11
  1

• Use amp and ImportError

  #48 opened 8 months ago by zzzyzh
  1

• can I use the sam_vit_h_4b8939.pth directly?

  #37 opened 10 months ago by Ghy1209
  3

• The data size after unzip

  #50 opened 8 months ago by YiyiMiao
  1

• Figure 4. in the Paper.

  #52 opened 7 months ago by ariharasudhanm
  0

• How long will it take to unzip full.zip?

  #40 opened 10 months ago by SVT-Yang
  1

• Hello, can Med2D train with multiple cards now? I failed to refer to the 3D multi card training.

  #49 opened 8 months ago by biyuefeng
  0

• How to train on multi gpus?

  #34 opened 10 months ago by KeyaoZhao
  2

• data

  #47 opened 8 months ago by biyuefeng
  3

• Dataset Download Unzip Problem

  #33 opened 10 months ago by Will-Zhao0
  2

• UnboundLocalError: local variable 'label' referenced before assignment

  #39 opened 10 months ago by Marsyule
  1

• The code for visualizing the results

  #44 opened 9 months ago by Lzy-dot
  1

• How to resize images to 256*256

  #41 opened 9 months ago by Devil-Ideal
  2

• 拥有空的或小的前景类区域的mask无法加载

  #38 opened 10 months ago by Marsyule
  1

• mask_decoder mismatch with raw SAM mask_decoder

  #42 opened 9 months ago by wade0604
  2

• Question about Training Set

  #43 opened 9 months ago by Huster-Hq
  1

• download dataset error

  #36 opened 10 months ago by w-sunmoon
  1

• Multiple boxes prompts

  #30 opened 10 months ago by Huster-Hq
  2

• SA-Med2D-20M Dataset Download Link

  #31 opened 10 months ago by MarkWijkhuizen
  2

• train

  #29 opened a year ago by closertt
  1

• WHY?

  #27 opened a year ago by GitHub-TXZ
  0

• After fine-tuning SAM with my own data, the model becomes very insensitive to prompt inputs.

  #25 opened a year ago by klykq111
  1

• Why does the training use BGR input instead of the original SAM's RGB input?

  #20 opened a year ago by klykq111
  2