Issues

• Why SegformerDecoder only uses 3 feature stages instead of 4 as in the original SegFormer paper?

  qubvel-org/segmentation_models.pytorch#1437 · Filter by author liuye23333 opened on Oct 10, 2026

• DPTDecoder doesn't work properly with VIT using patch_size=14

  qubvel-org/segmentation_models.pytorch#1328 · Filter by author schwobr opened on Apr 20, 2026

• The training result is so much lower than the sample code.

  qubvel-org/segmentation_models.pytorch#1323 · Filter by author FreemanTang opened on Apr 13, 2026
  2

• ONNX Unsupported export of operator adaptive_avg_pool2d

  qubvel-org/segmentation_models.pytorch#1266 · Filter by author statiqueplasma opened on Feb 10, 2026

• Inconsistent from_logits parameter across loss functions

  qubvel-org/segmentation_models.pytorch#1263 · Filter by author AndrewFalkowski opened on Dec 2, 2025
  3

• How to load Huggingface pretrained model for SMP training?

  qubvel-org/segmentation_models.pytorch#1252 · Filter by author james-imi opened on Oct 14, 2025

• can I use convnext to unet model

  qubvel-org/segmentation_models.pytorch#1251 · Filter by author pure-rgb opened on Oct 14, 2025

• VIT Adapter

  qubvel-org/segmentation_models.pytorch#1244 · Filter by author aselimc opened on Sep 21, 2025
  5

• how to print Dice,HD95 metrics?

  qubvel-org/segmentation_models.pytorch#1241 · Filter by author zhaoyang23333 opened on Sep 18, 2025

• SOTA Panoptic Methods Implementation

  qubvel-org/segmentation_models.pytorch#1240 · Filter by author aselimc opened on Sep 18, 2025

• segformer model implementation != original arch design

  qubvel-org/segmentation_models.pytorch#1237 · Filter by author pure-rgb opened on Sep 15, 2025
  8

• Possibility of optimisation in Focal Loss implementation

  qubvel-org/segmentation_models.pytorch#1235 · Filter by author vedantdalimkar opened on Sep 14, 2025
  1

• Example "Train multiclass segmentation on CamVid" Image normalization may be error

  qubvel-org/segmentation_models.pytorch#1224 · Filter by author sysau opened on Aug 29, 2025
  2

• tu-vit_base_patch16_224 not supported

  qubvel-org/segmentation_models.pytorch#1184 · Filter by author isaaccorley opened on Jun 22, 2025
  4

• AssertionError description typo

  qubvel-org/segmentation_models.pytorch#1167 · Filter by author goodsuga opened on May 28, 2025

• KeyError when unet uses certain encoders

  qubvel-org/segmentation_models.pytorch#1159 · Filter by author jdg325 opened on May 26, 2025
  3

• Changing dilation rate of backbones

  qubvel-org/segmentation_models.pytorch#1158 · Filter by author vedantdalimkar opened on May 23, 2025
  1

• Exporting UnetPlusPlus as TorchScript gives error

  qubvel-org/segmentation_models.pytorch#1156 · Filter by author atncai opened on May 22, 2025

• How to migrate existing models from 0.4.0 to 0.5.0?

  qubvel-org/segmentation_models.pytorch#1144 · Filter by author ChristianWiele opened on May 5, 2025
  4

• Add EoMT from ViT is Secretly an Image Segmentation Model

  good difficult issuenew-model
  qubvel-org/segmentation_models.pytorch#1132 · Filter by author tcourat opened on Apr 19, 2025
  2

• Addition of other losses

  qubvel-org/segmentation_models.pytorch#1114 · Filter by author GuillaumeErhard opened on Apr 3, 2025

• Input Shape for Training and Inference with Pretrained Models

  qubvel-org/segmentation_models.pytorch#1112 · Filter by author hamac03 opened on Apr 2, 2025
  2

• Add upsampling parameter to Segformer

  featuregood first issue
  qubvel-org/segmentation_models.pytorch#1106 · Filter by author thierrypin opened on Mar 26, 2025
  3

• Encoder weights from local directory

  qubvel-org/segmentation_models.pytorch#1102 · Filter by author Foundsheep opened on Mar 25, 2025
  4

• Is there any other pre-trained model in this pkg besides segformer?

  qubvel-org/segmentation_models.pytorch#1100 · Filter by author yunruo-zhang-will opened on Mar 24, 2025
  2