mit-han-lab/efficientvit

Issues

• Adding a training pipeline for the DC-AE (Deep Compression Autoencoder train pipeline)

  #148 opened 22 days ago by zeroshot-ai
  3

• Training hyper-parameters for DCAE

  #146 opened 25 days ago by francois-rozet
  0

• Why the FID of Flux-VAE-f8 is worse than SD-VAE-f8 in table 3 ?

  #147 opened 24 days ago by cbstyle
  0

• 是否可以直接替换stable diffusion的vae使用

  #145 opened a month ago by catsled
  1

• 模型编码器tensorrt测试是用的fp16模式吗

  #142 opened a month ago by 13213085
  1

• What does in/mix mean?

  #144 opened a month ago by Godofnothing
  5

• Question about DC AE

  #143 opened a month ago by MaxxP0
  2

• Deprecated Usage of torch.cuda.amp.autocast

  #127 opened a month ago by pg56714
  0

• Set weights_only=True in torch.load for security

  #132 opened a month ago by pg56714
  0

• Questions about DC-AE

  #141 opened a month ago by theFoxofSky
  4

• There are some problems with this website：https://evitsam.hanlab.ai/

  #119 opened 5 months ago by LHX1220328779
  5

• Diffusion DC-AE example errors - AssertionError: fid ref assets/data/fid/imagenet_512_train.npz not existed

  #139 opened a month ago by rockerBOO
  3

• Running very slowly

  #136 opened 2 months ago by CurryKd7
  0

• trt后如何使用全图分割

  #135 opened 2 months ago by xiaodu0414
  0

• Hyperparameter request for reproducibility

  #134 opened 3 months ago by ovunctuzel-bc
  2

• NaN values with FP16 TensorRT Inference

  #116 opened 6 months ago by ovunctuzel-bc
  6

• TensorRT Inference for Segmentation Models

  #130 opened 3 months ago by Sanath1998
  0

• Unmatch between code and paper

  #129 opened 4 months ago by Rbrq03
  0

• sam inference using onnx is much slower than trt inference

  #126 opened 4 months ago by superran1
  0

• Excuse me, may I ask When will the Segmentation training code be updated

  #120 opened 5 months ago by 111hyq111
  1

• where is EfficientViTBlock？

  #125 opened 4 months ago by saccy6
  0

• why the train loss without iou_prediction?

  #124 opened 4 months ago by cvhuang
  0

• convert model to TensorRT10

  #123 opened 4 months ago by yflv-yanxia
  0

• Request for Code Release: Condition-Aware Neural Network for Controlled Image Generation

  #98 opened 7 months ago by MikePrivateCoding
  3

• Nice

  #122 opened 4 months ago by Static3dw
  0

• Is there any choice for segment the salient region on the image?

  #121 opened 5 months ago by yanxisusu
  0

• SAM box issue using mps (macos)

  #105 opened 6 months ago by Chris-toe-pher
  1

• Regarding the training code for SAM

  #118 opened 5 months ago by feivelliu
  0

• ONNX inference for MaskGenerator

  #117 opened 5 months ago by yujunsun110
  0

• Discrepancy between TRT engines for same models - TensorRT Issue cross-reference

  #113 opened 6 months ago by bernardrb
  1

• Request: EfficientViT for CLIP

  #115 opened 6 months ago by rotem154154
  0

• FFN+DWConv in paper vs code

  #114 opened 6 months ago by seermer
  1

• [Semantic segmentation output size problem]

  #111 opened 6 months ago by CaptainEven
  1

• When wiil you release the train code of seg?

  #109 opened 6 months ago by Anivie
  1

• When will the code for semantic segmentation be publicly available?

  #107 opened 6 months ago by ascacdsaa
  1

• Pretrain on ImageNet-21K dataset

  #112 opened 6 months ago by tw0b33rs
  0

• FP16 training will occurs NaN problem after last code updates

  #110 opened 6 months ago by xyzacademic
  3

• ONNX only supports point prompts and box prompts. How to segment everything with onnxruntime?

  #104 opened 6 months ago by panp4n
  2

• Custom dataset finetuning

  #108 opened 6 months ago by Ostyk
  0

• Segmentation scores are wrong

  #103 opened 7 months ago by franchesoni
  0

• train the image encoder of EfficientViT-SAM

  #102 opened 7 months ago by Yangenhui98
  0

• What is the difference between efficient vit and efficient sam?

  #101 opened 7 months ago by xiawei20161308104
  0

• Could someone explain to me the operation of efficient attention？

  #100 opened 7 months ago by CHDyshli
  0

• Request for releasing the segcore module for segmentation training

  #99 opened 7 months ago by carvychen
  0

• what's the version of tensorrt?

  #96 opened 7 months ago by freeman745
  2

• when to release the training code of EfficientViT-SAM?

  #94 opened 8 months ago by Dongshengjiang
  1

• A question about EfficientViTBlock and EfficientViTModule in the paper

  #97 opened 7 months ago by xiawei20161308104
  0

• 1x1 group convs in LiteMLA

  #95 opened 8 months ago by benjamindorra
  0

• The output from c++_trt infer is not same as the py_trt when i set two or more points to segment

  #93 opened 8 months ago by yangchengxin
  1

• The output from c++_trt Infer

  #92 opened 8 months ago by yangchengxin
  0