ExponentialML/Text-To-Video-Finetuning

Issues

• init_video problem

  #153 opened a year ago by Yuuki-00
  0

• TypeError: Linear.forward() got an unexpected keyword argument 'scale'

  #134 opened a year ago by kenkenissocool
  6

• ControlNet

  #150 opened a year ago by julkaztwittera
  0

• Normal finetuning instead of LoRA

  #149 opened a year ago by julkaztwittera
  0

• About step_loss of version2

  #46 opened 2 years ago by suzhenghang
  4

• issues on train.py

  #136 opened a year ago by MuhammadNaeem42
  1

• Lora on ResnetBlock2D in modelscope model

  #142 opened a year ago by joaanna
  1

• Can the videocomposer model be adapted to this training framework?

  #144 opened a year ago by YujieOuO
  0

• wrong norm method

  #135 opened a year ago by BooHwang
  1

• Two forward passes in finetune_unet

  #139 opened a year ago by gunshi
  1

• [inference] latents_window index error

  #138 opened a year ago by IceTea42
  1

• Enabling Multi-GPU training

  #73 opened a year ago by julkaztwittera
  3

• Lora inference problem

  #78 opened a year ago by zacaikido
  7

• A typo

  #132 opened a year ago by pixeli99
  0

• AttributeError: 'Tensor' object has no attribute 'config'

  #106 opened a year ago by howardgriffin
  5

• How can I run the fine-tuning on a GPU with <= 16GB of VRAM?

  #118 opened a year ago by mmakiuchi
  3

• I have some doubts about the framework

  #125 opened a year ago by pixeli99
  4

• RuntimeError: cannot reshape tensor of 0 elements into shape [0, -1, 1, 512] because the unspecified dimension size -1 can be any value and is ambiguous

  #104 opened a year ago by Splendon
  3

• Does this code support native finetune for damo text to video model?

  #105 opened a year ago by ersanliqiao
  2

• Which paper?

  #103 opened a year ago by howardgriffin
  1

• How to train with folder video

  #102 opened a year ago by Splendon
  1

• webui Lora Might be causing errors in checkpoint models.

  #101 opened a year ago by justinwking
  3

• How to test my model

  #91 opened a year ago by Revanthraja
  9

• finetune train error of "UnboundLocalError: local variable 'use_offset_noise' referenced before assignment"

  #99 opened a year ago by Splendon
  2

• error of "from accelerate.logging import get_logger"

  #98 opened a year ago by Splendon
  2

• Transformer2D initializing

  #82 opened a year ago by johnmullan
  1

• training video

  #92 opened a year ago by cherfffff
  0

• TypeError: UNet3DConditionModel._set_gradient_checkpointing() got multiple values for argument 'value'

  #69 opened a year ago by FurkanGozukara
  3

• How much VRAM do I need for saving the weights?

  #72 opened a year ago by Darkbblue
  4

• Refer to the official release of Diffusers?

  #77 opened a year ago by tin2tin
  2

• similar implementation to Nivida VideoLDM?

  #79 opened a year ago by Maki9009
  5

• Accelerator 'function' object has no attribute '__func__'

  #83 opened a year ago by justinwking
  3

• Question: regarding fine-tuning with images

  #59 opened 2 years ago by patrickjonesdotca
  1

• About VideoLDM

  #55 opened 2 years ago by suzhenghang
  7

• Feature request

  #57 opened 2 years ago by justinwking
  9

• Encountered 2 file(s) that may not have been copied correctly on Windows

  #60 opened a year ago by 1-sf
  1

• Default model seems to output only noise or greenscreen

  #62 opened 2 years ago by patrickjonesdotca
  6

• First GPU occupies more VRAM in distributed training

  #66 opened a year ago by suzhenghang
  0

• inference code generate only green noise

  #74 opened a year ago by ChrisTiger93
  1

• Links to weights?

  #54 opened a year ago by tin2tin
  1

• about vid2vid in inference

  #64 opened 2 years ago by suzhenghang
  1

• Error with inference

  #61 opened 2 years ago by dvschultz
  2

• NameError: name 'glob' is not defined

  #45 opened 2 years ago by ImBadAtNames2019
  40

• model inference of version2

  #50 opened 2 years ago by suzhenghang
  7

• Main ways to avoid overfitting?

  #44 opened 2 years ago by ImBadAtNames2019
  1

• Whats the deal with train_text_encoder: ?

  #43 opened 2 years ago by ImBadAtNames2019
  1

• Is there a problem with attn3 here?

  #40 opened 2 years ago by suzhenghang
  1

• How does n_sample_frames: work?

  #42 opened 2 years ago by ImBadAtNames2019
  1

• Question about finetuning multiple videos, each one with a different prompt, using the video_folder.yaml

  #41 opened 2 years ago by ImBadAtNames2019
  2

• How do you turn off validation?

  #38 opened 2 years ago by ImBadAtNames2019
  1