baaivision/EVA

Issues

• About SigLIP

  #166 opened 4 months ago by ApocalypsezZ
  0

• I would like to know the text input capacity of eva-clip-18b? To my knowledge, OpenAI's CLIP requires less than 20 tokens / 我想了解一下 eva-clip-18b 文本输入容量是多少？据我了解，OpenAI 的 CLIP 的话低于20 个 token

  #165 opened 5 months ago by gg22mm
  0

• What's the difference with metric.json and log.txt

  #163 opened 5 months ago by EunNamCho
  0

• issue about “eot_token”

  #159 opened 6 months ago by konioy
  0

• AttributeError: module 'PIL.Image' has no attribute 'LINEAR'. Did you mean: 'BILINEAR'?

  #162 opened 6 months ago by dgcnz
  0

• Clarification about EVA-CLIP / EVA-02-CLIP

  #161 opened 6 months ago by matteot11
  0

• Size 14x14 to 16x16 patch interpolation for smaller EVA 2 models

  #153 opened 9 months ago by tobiasvanderwerff
  2

• [eva02 detection]why need to interpolate pos_emb as 14*14 .

  #160 opened 6 months ago by h-bo
  0

• GFLOPs value

  #158 opened 7 months ago by ppriyank
  0

• 我在用eva02，加载pretrainedmodel的时候"rope" size mismatch，我看您的mmcv_custom里load_checkpoint方法没有对这一mismatch进行处理

  #157 opened 7 months ago by yan-hao-tian
  0

• deepspeed zero1 not support bf16

  #155 opened 7 months ago by xiaomabufei
  0

• Where is EVA G?

  #152 opened 8 months ago by OpenJarvisAI
  3

• Chinese ability of EVA-CLIP-18B

  #154 opened 8 months ago by Limingxing00
  0

• EVA CLIP 8B GPU out of memory in COLAB PRO Hugging face version

  #147 opened 9 months ago by chandrabhuma
  1

• can remove xformers dependency?

  #148 opened 9 months ago by zimenglan-sysu-512
  1

• How much GPU does EVA-CLIP-8B-plus need?

  #151 opened 9 months ago by ontheway-arch
  0

• Dockerfile error

  #150 opened 9 months ago by guozj583
  0

• Is there any reason why you didn't use the EVA02 architecture for EVA-CLIP models larger than 4B?

  #144 opened 9 months ago by stevenliu000
  1

• different weight decay between code and paper for clip 18b

  #145 opened 9 months ago by Novestars
  1

• [EVA-CLIP] Are you sure about your way to compute probabilities?

  #140 opened 9 months ago by paulgavrikov
  1

• Can EVA-CLIP-8B and EVA-CLIP-18B support quantization?

  #142 opened 9 months ago by Shimooth
  6

• Issues about the performance of vanilla EVA-02-S

  #149 opened 10 months ago by jwfanDL
  0

• How to download the init_weight for seg model, the link is faild?

  #120 opened a year ago by zhangyan666
  2

• Any plans for releasing model weights of COCO-Stuff trained segmentation models with EVA-02 backbones

  #146 opened 10 months ago by ankuPRK
  0

• Why are the model weights saved at each epoch the same when using DeepSpeed?

  #141 opened 10 months ago by konioy
  1

• ```EVA-CLIP-18B``` weight is not available at HuggingFace

  #143 opened 10 months ago by shyammarjit
  2

• Is there any plan to use lora for finetune eva_clip?

  #138 opened a year ago by dongzhiwu
  1

• How to use EVA-02 detection models for inference?

  #139 opened a year ago by Franklin905
  0

• whether the training of EVA involves masking text (caption) token?

  #135 opened a year ago by leyangjin
  1

• How to run Inference and Fine-Tuning on Instance Segmentation EVA-02?

  #127 opened a year ago by FrancoisPorcher
  0

• Robust accuracy on ObjectNet dataset.

  #136 opened a year ago by sunrise6513
  0

• CUDA out of memory when inference

  #134 opened a year ago by zhaozhen2333
  0

• About export model

  #117 opened a year ago by Egorundel
  8

• Under the same validation set, the evaluation results of the EMA model after loading are inconsistent with those during training.

  #133 opened a year ago by FriedaSmith
  1

• use torch.jit.trace export pytorch 2 torchscript fail.

  #131 opened a year ago by xinlin-xiao
  8

• Training loss is unstable

  #125 opened a year ago by Kizna1ver
  5

• Can eva_video_k722 be used to extract video clip features?

  #132 opened a year ago by PolarisHsu
  0

• How to convert class id to class name

  #129 opened a year ago by huijiang2486
  1

• Pooling strategy

  #130 opened a year ago by mactavish91
  0

• how to get the hidden features by api code?

  #126 opened a year ago by hardlipay
  1

• Ripple-like Artifacts on Mask Surfaces with EVA Model

  #128 opened a year ago by asdlkjw
  0

• Cannot find Evaluation Metrics on LVIS

  #123 opened a year ago by FrancoisPorcher
  2

• cannot run the detection training with mutiple gpu on a single node

  #122 opened a year ago by lixi92
  3

• About post normalization

  #124 opened a year ago by HJYao00
  0

• Issue with Detectron2 and CUDA

  #121 opened a year ago by FrancoisPorcher
  0

• The import accimage in the dateset folder cannot resolve the load.

  #119 opened a year ago by zhanglaoban-kk
  0

• the .sh script of Evaluate the fine-tuned EVA (336px, patch_size=14) on ImageNet-1K val with a single node (click to expand) can not execute.

  #118 opened a year ago by peter-ni-noob
  1

• EVA-01 and EVA-02

  #116 opened a year ago by Egorundel
  0

• Error when loading&training EVAVisionTransformer with patch_dropout=0.5

  #115 opened a year ago by chenmc1996
  0

• deepspeed can not work

  #114 opened a year ago by shuowoshishui
  0