google-research/big_vision

Issues

• Should SigLip memory consumption increase as we scale number of GPUs

  #132 opened 3 months ago by khalidsaifullaah
  0

• GIVT (and more generally, new research by Google) on Hugging Face

  #131 opened 3 months ago by NielsRogge
  0

• Add chunked implementation of siglip sigmoid loss

  #130 opened 3 months ago by cdeepakroy
  0

• [Question] How to inference / captioning short video?

  #117 opened 6 months ago by yoesak
  5

• Causal Mask leads to NaN gradients (Paligemma)

  #129 opened 4 months ago by sanjayss34
  1

• Loss Scale for Training Siglip

  #115 opened 6 months ago by lezhang7
  1

• tokenization error when using msiglip

  #126 opened 4 months ago by simran-khanuja
  1

• What's a "`rezied` method"?

  #124 opened 5 months ago by EIFY
  0

• Could you please give me an instruction on how to use paligemma's evaluator?

  #123 opened 5 months ago by handsomelys
  0

• Question about the evaluation on instance segmentation about PaliGemma

  #118 opened 6 months ago by handsomelys
  0

• PlaiGemma finetuned model

  #113 opened 6 months ago by TheMattBin
  1

• Can I convert paligemma npz model to pytorch to safetensors?

  #107 opened 7 months ago by sdip15fa
  3

• No ChartQA ft weights released on Kaggle

  #111 opened 7 months ago by probicheaux
  1

• How to save fine tuned PaliGemma model?

  #104 opened 7 months ago by SkalskiP
  3

• Behavior of `solarize()` depends on integer overflow

  #110 opened 7 months ago by EIFY
  0

• [QUESTION] How to perform inference on trained model?

  #106 opened 7 months ago by sleepingcat4
  8

• Implementation of contrast() seems wrong

  #109 opened 7 months ago by EIFY
  0

• [BUG] in big_vision.models.proj.flexi.vit

  #76 opened 7 months ago by KennyFu1997
  1

• Error with putting arrays on CPU in cloud TPUs

  #101 opened 8 months ago by philippe-eecs
  1

• Confusion on FlexiViT

  #94 opened 10 months ago by zilunzhang
  0

• Errors in notebooks

  #93 opened 10 months ago by amrzv
  0

• Clarification: SigLIP Image Transform

  #91 opened a year ago by siddk
  0

• Mixup Per Example?

  #90 opened a year ago by dibyaghosh
  0

• Load ViT with CLIPPO Weights

  #88 opened a year ago by ahmadmustafaanis
  2

• Reproduced result for flexivit

  #77 opened a year ago by liguopeng0923
  2

• FlexiVit is also flexible with image resolution?

  #86 opened a year ago by KennyFu1997
  0

• Negative rho values in GSAM training

  #80 opened a year ago by ankitkv
  1

• Text lowering issue

  #79 opened a year ago by shkarupa-alex
  0

• SigLIP and canonicalize

  #78 opened a year ago by shkarupa-alex
  0

• Contrastive Input Pipeline

  #75 opened a year ago by philippe-eecs
  2

• Question about SigLIP

  #58 opened a year ago by maximzubkov
  7

• Could you provide the checkpoint of the CLIPPO model?

  #29 opened 2 years ago by zzhanghub
  11

• Question About Listed ViT Models in the configs/proj/flexivit/README.md

  #31 opened 2 years ago by mhamzaerol
  2

• Question regarding value range (-1,1)

  #37 opened a year ago by LeoXinhaoLee
  6

• Running out of RAM on cloud TPU when reading data from Cloud Storage

  #36 opened a year ago by izmailovpavel
  2

• Announcement: big_vision is transitioning from jax.pmap to jax.jit.

  #52 opened a year ago by akolesnikoff
  1

• Is there Pytorch version CLIPPO?

  #57 opened a year ago by yukistavailable
  2

• bfloat16 Training

  #41 opened a year ago by LeoXinhaoLee
  1

• question about FlexiViT

  #30 opened a year ago by KimWu1994
  1

• questions about t-SNE visualization in FlexiViT

  #43 opened a year ago by ShipengFu
  1

• Question about ViT-augreg ("How to train?") fine-tuning transfer

  #60 opened a year ago by lucasb-eyer
  2

• Memory Efficient Attention integration

  #59 opened a year ago by maximzubkov
  2

• Question: Updating mask in classification evaluator

  #42 opened a year ago by MasterSkepticista
  1

• Optax update breaks due to jax.Array migration

  #35 opened a year ago by colbybanbury
  3

• Gradient accumulation

  #54 opened a year ago by Mahesh-Ram1
  0

• Question: Will SigLIP / SigLiT be added to this codebase?

  #38 opened a year ago by linzhiqiu
  5

• requirements issue

  #49 opened a year ago by sneccc
  2

• Add other custom optimizers

  #32 opened 2 years ago by buttercutter
  2

• About RL fine-tuning code release

  #40 opened a year ago by byminji
  2

• Colorization uvim model not working

  #19 opened 2 years ago by muhammad-ahmed-ghani
  5