NVlabs/FasterViT

Issues

• import error

  #144 opened 3 months ago by MohammedAbusarie
  1

• How to determine the args.mesa ratio?

  #143 opened 5 months ago by Kim-DKyu
  0

• ImageNet-21K checkpoint without fine-tuning!

  #137 opened 5 months ago by black0017
  2

• Mismatch with timm library

  #141 opened 7 months ago by MaratZakirov
  1

• Why use CNN in the first two stage.

  #138 opened 7 months ago by jamesben6688
  2

• HAT

  #126 opened 7 months ago by YSLDTZY
  1

• Training FasterViT on VisDrone2019 dataset

  #125 opened 7 months ago by bshakya77
  1

• Latency under bs=1 seems not impressive like throughput under bs=128.

  #129 opened 7 months ago by TsingWei
  2

• may be upload wrong ckpts with faster_vit_0_any_res?

  #139 opened 7 months ago by XiaobinWu1998
  0

• Model Evaluation

  #127 opened 8 months ago by bshakya77
  1

• Training Issue

  #130 opened 8 months ago by bshakya77
  1

• ImportError: cannot import name

  #136 opened 8 months ago by 0xrsydn
  1

• Drawing Heatmap

  #135 opened 8 months ago by kojunseo
  0

• TokenInitializer for any resolution

  #78 opened a year ago by becauseofAI
  2

• confusion about ct_dewindow

  #79 opened a year ago by becauseofAI
  2

• any_ res is not truly any resolution

  #76 opened a year ago by becauseofAI
  5

• Object Detection with FasterViT-0 an 1

  #26 opened a year ago by chuong98
  2

• Inquiry for detection codes

  #119 opened a year ago by George0726
  1

• level-dimension

  #55 opened a year ago by houph4
  2

• Detection and segmentation models

  #21 opened a year ago by pianogGG
  8

• Size mismatch when loading pretrained weights.

  #118 opened a year ago by TsingWei
  1

• How to extract features from images of different sizes

  #116 opened a year ago by huangzhufan
  3

• Use FasterViT model like a backbone for segmentation model

  #114 opened a year ago by heorhiikalaichev
  1

• ONNX simplify error occurs when either width or height is not a multiple of 7

  #106 opened a year ago by chohk88
  1

• How to profile the DL training and inference

  #110 opened a year ago by falcon-xu
  1

• About model profiling.

  #75 opened a year ago by TsingWei
  2

• Cross- attention

  #98 opened a year ago by williamhoole
  0

• Issue Converting FasterViT to CoreML

  #91 opened a year ago by rotem154154
  8

• Plug and Play HAT

  #77 opened a year ago by williamhoole
  0

• Help

  #80 opened a year ago by zbl929
  1

• Bug in fastervit package: faster_vit_2_any_res

  #72 opened a year ago by luke-mcdermott-mi
  1

• How to optimize the performance of Fastervit's int8

  #71 opened a year ago by tp-nan
  1

• Improved fastervit_any_res_0 has larger TensorRT latency than the original version

  #62 opened a year ago by QingXIA233
  9

• Issue with faster_vit_4_224 checkpoint

  #56 opened a year ago by numpee
  7

• Allow for running HAT via seperate flag

  #59 opened a year ago by ahatamiz
  1

• FSDP causes graph break in forward

  #51 opened a year ago by Gutianpei
  3

• ModuleNotFoundError: No module named 'utils'

  #50 opened a year ago by 18370636086
  1

• model in deepcopy will get error.

  #24 opened a year ago by z1069614715
  12

• confuse about use pretrained model to finetune on 1024 x 1024 images

  #43 opened a year ago by dongshenhan
  1

• Loading pre-trained model

  #31 opened a year ago by Maha59
  4

• different resolution between train and inference

  #22 opened a year ago by d-zhou12
  4

• same gamma parameter

  #35 opened a year ago by OnderT
  1

• Window size and ct_size setting when resolution_h doesn't equal resolution_w

  #3 opened a year ago by QingXIA233
  6

• Error while exporting to dynamic onnx

  #12 opened a year ago by tp-nan
  8

• Paper error

  #23 opened a year ago by zbl929
  1

• Wrong shape order in TokenInitailizer?

  #6 opened a year ago by Birdylx
  3

• export onnx error

  #2 opened a year ago by mtjhl
  2

• When will the paper be published?

  #1 opened a year ago by terafo
  2