pixeli99/SVD_Xtend

Issues

• Tensor size mismatch error

  #33 opened 9 months ago by SwayStar123
  7

• RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 6 but got size 5 for tensor number 1 in the list.

  #53 opened 7 months ago by TIANTEA
  4

• 36gb minimum GPU memory is required using batch size 1 and fp16 mixed precision training?

  #61 opened 4 months ago by danielvegamyhre
  9

• About the GPU settings when you training.

  #31 opened 10 months ago by yunniw2001
  17

• pretrained_model_name_or_path报错

  #52 opened 4 months ago by 1585511010
  2

• Provide requirements.txt by running `pip3 freeze > requirements.txt` from a working development environment

  #59 opened 4 months ago by danielvegamyhre
  1

• 请问下BF16是怎么执行的，我执行有报错，帮看下

  #51 opened 4 months ago by zhanghaobucunzai
  1

• Which BDD100K data is used for training?

  #20 opened 5 months ago by canqin001
  2

• text2video data

  #36 opened 5 months ago by HaoZhang990127
  1

• Question about the forward pass

  #32 opened 5 months ago by KKN18
  1

• Maybe there's a mistake with the value range when sending the image into the VAE.

  #16 opened 5 months ago by bugWholesaler
  3

• Questions about the predicted_noise of EDM framework.

  #19 opened 5 months ago by Pandint
  3

• Training on multi GPUs with different memory size

  #58 opened 5 months ago by KhaledButainy
  0

• How to support batch_size > 1?

  #49 opened 7 months ago by liiiiiiiiil
  1

• after training on 512x512, the video not move always, why?

  #56 opened 6 months ago by ersanliqiao
  2

• About the step_loss == nan

  #17 opened 10 months ago by maobenz
  7

• [CUDA out of memory] training in 1024 × 576 resolution in the A100 80G

  #39 opened 9 months ago by CallMeFrozenBanana
  4

• 多卡训练报错

  #42 opened 8 months ago by KaiGod0730
  2

• Question about classifier guidance for image in training code

  #54 opened 6 months ago by tykim0507
  3

• Fail to start the training at high resolutions.

  #29 opened 10 months ago by Little-Podi
  4

• Exception: Could not find the transformer layer class to wrap in the model.

  #55 opened 6 months ago by matbee-eth
  0

• 能否提供下pretrained_model_name_or_path的下载路径？

  #37 opened 9 months ago by zhanghaobucunzai
  3

• May I ask what pretrained_model_name_or_path is

  #13 opened a year ago by aoyingying
  4

• How can I use the trained module

  #48 opened 7 months ago by fyb0101
  0

• Add a short inference script loading from checkpoint?

  #47 opened 7 months ago by nqyy
  3

• Error when I start the training

  #46 opened 8 months ago by andreichisilev
  2

• losses barely dropped with BDDx datasets？？

  #45 opened 8 months ago by qiuhaining
  0

• Thank you, fine_tuned weights

  #43 opened 8 months ago by wisleythewise
  0

• Suggestions on loading video data

  #41 opened 8 months ago by Semolce9
  1

• Question：How to use the trained modules to generate a video？

  #40 opened 9 months ago by Henrik-hw
  0

• Finetune SVD with discrete time noise scheduler

  #35 opened 9 months ago by HL-hanlin
  2

• WARNING - tensorboardX.x2num - NaN or Inf found in input tensor.

  #34 opened 9 months ago by Henrik-hw
  1

• Questions on text2video?

  #25 opened 10 months ago by hitsz-zuoqi
  5

• Multi-gpu training

  #24 opened 10 months ago by KKN18
  3

• Question for the encoder_hidden_states

  #28 opened 10 months ago by WayneML
  4

• DeepSpeed Accelerate Config

  #30 opened 10 months ago by zeyuanchen23
  0

• Out of Memory

  #27 opened 10 months ago by AbrahamYabo
  2

• Question about the new push for partial fine-tuning.

  #26 opened 10 months ago by Kiteretsu77
  2

• multi-gpu training?

  #23 opened 10 months ago by yanghbcodes
  0

• Questions about the noise sampling.

  #21 opened 10 months ago by Muon2
  10

• What can I change if I want to add LoRA method into your train script ?

  #22 opened 10 months ago by WayneML
  1

• Questions about the noise sampling.

  #18 opened a year ago by Pandint
  2

• just thank you for share this SVD training code

  #8 opened a year ago by shamu899063
  2

• How much memory is needed when training SVD

  #9 opened a year ago by dailingx
  1

• Support more conditional inputs, such as layout

  #10 opened a year ago by zhw-zhang
  3

• thank you and do you use the frames at 10th second for training?

  #15 opened a year ago by hitsz-zuoqi
  2

• Do you have any advice on text2video?

  #11 opened a year ago by Len-Li
  2

• AttributeError: 'UNetSpatioTemporalConditionModel' object has no attribute 'module'. Did you mean: 'modules'?

  #12 opened a year ago by dailingx
  2

• Hyperparameters configuration for training

  #6 opened a year ago by MyNiuuu
  2

• training gpu cost

  #4 opened a year ago by zhw-zhang
  3