czczup/ViT-Adapter

Issues

• 从 mmseg0.x 迁移至 mmseg 1.x 复现出现问题

  #160 opened a year ago by AI-Tianlong
  3

• Utilize lightweight detection head

  #188 opened a month ago by King4819
  0

• Using ssd300 as detector

  #187 opened a month ago by King4819
  0

• Poor performance on custom dataset.

  #155 opened a year ago by lianzheng-research
  1

• Utilize yolo detection head

  #186 opened a month ago by King4819
  0

• Drop_path_rate rule

  #185 opened 2 months ago by iacopo97
  0

• 作者团队好，请问如果没有injector，只有extractor，性能表现怎么样？我看论文里的消融实验（Table 6）好像没有提及这个，有做过这方面对比吗

  #182 opened 2 months ago by yan-hao-tian
  0

• img_norm_config for maskrcnn dinov2 large seems to be accidentally default

  #184 opened 3 months ago by CardenB
  2

• mmcv and mmsegmentation versions incompatability in Docker environment

  #183 opened 3 months ago by kaavyyaashri
  0

• how to deploy using TensorRT?

  #180 opened 4 months ago by chenzhutian
  1

• Wanted to know if there is an MAE based implementation for Vit Adapter ?

  #181 opened 3 months ago by ayushnangia
  0

• KeyError: 'ViTAdapter is not in the models registry'

  #154 opened a year ago by CatOneTwo
  3

• Checkpoint cannot download

  #179 opened 4 months ago by LiGuo12
  0

• CUDA and Pytorch version mismatch

  #174 opened 6 months ago by arslanduygu
  2

• Issues with Semantic Segmentation on custom datasets

  #175 opened 5 months ago by HarshitSheoran
  2

• Performance of shared_rel_pos_bias

  #177 opened 5 months ago by bio-mlhui
  0

• Code Problem: not consistent with original BEiT implementation

  #178 opened 5 months ago by bio-mlhui
  0

• Poor performance of Mask2Former based on coco-stuff-164k

  #141 opened a year ago by clxia12
  6

• ModuleNotFoundError: No module named 'MultiScaleDeformableAttention'

  #176 opened 5 months ago by Githia
  0

• Training steps on semantic segmentation

  #173 opened 6 months ago by yuyu19970716
  0

• Instance Segmentation Model Not Covering Entire Image

  #172 opened 6 months ago by SaulZhang
  0

• error in CUDA

  #171 opened 6 months ago by iacopo97
  0

• get_flops.py

  #168 opened 8 months ago by Minsu-Kim1822
  11

• TypeError: forward_train() missing 2 required positional arguments: 'gt_labels' and 'gt_masks'

  #140 opened a year ago by SaulZhang
  5

• reference point visualization and some questions

  #170 opened 8 months ago by SnaKey0u0
  0

• "MaskRCNN: 'ViTBaseline is not in the models registry'"

  #169 opened 8 months ago by LQchen1
  0

• 关于drop path rate的问题

  #167 opened 8 months ago by Wenlve-Zhou
  1

• ModuleNotFoundError: No module named 'MultiScaleDeformableAttention'

  #166 opened 9 months ago by MF-XU
  5

• After Freezing the Vit backbone how many iterations are supposed to be trained?

  #147 opened a year ago by iacopo97
  1

• Gradient accumulation implemention

  #165 opened 9 months ago by King4819
  2

• Inference in batches and multiple GPU

  #163 opened 9 months ago by OmergottliebAB
  1

• 关于segmentation和detection的BEiT实现略有不同的问题

  #161 opened 10 months ago by LiJiahang617
  2

• Missing pretrained weights

  #162 opened 10 months ago by manu15sd
  1

• LayerScale and pre-trained models

  #157 opened 10 months ago by ds2268
  5

• Request the configuration file of objects365 pre-training

  #159 opened a year ago by AyanamiReiFan
  0

• May I ask how to export onnx? There has been an error

  #150 opened a year ago by wenshuaishuai123
  1

• Can ViT-Adapter be exported as an ONNX model? What are the respective versions of PyTorch and the ONNX subset for this?

  #158 opened a year ago by GitHubYuxiao
  0

• Dimension mismatching for beit_dapter with changing img_size and crop_size from (512, 512) to (384, 384)

  #156 opened a year ago by lianzheng-research
  0

• About the GFlops on the ADE20K dataset.

  #144 opened a year ago by SaulZhang
  3

• knowledge distillation

  #143 opened a year ago by GitHubYuxiao
  1

• Relative log amplitudes of Fourier transformed feature maps.

  #152 opened a year ago by Harry-zzh
  0

• Reccomended config for the case of binary segmentation

  #153 opened a year ago by iacopo97
  0

• ViT Adapter Not Working With Patch Size Different From 16

  #151 opened a year ago by MatCorr
  1

• number of iteration in the case of a batchsize equals to 1

  #149 opened a year ago by iacopo97
  0

• How can I use other pretrained ViTs or CNNs?

  #148 opened a year ago by lianzheng-research
  0

• Cannot download the checkpoint of upernet_deit_adapter_tiny.

  #146 opened a year ago by aluckyi
  0

• swin-t-adapter

  #145 opened a year ago by fyting
  1

• ModuleNotFoundError: No module named 'ops.modules'

  #142 opened a year ago by GitHubYuxiao
  4

• ImportError: cannot import name 'MultiScaleDeformableAttention' from 'mmcv.cnn.bricks.transformer'

  #139 opened a year ago by Coldwood1026
  1

• hi, I want to use this framework for visual grounding tasks, can you give me some suggestions to address this?

  #138 opened a year ago by RYHSmmc
  0