rinongal/textual_inversion issues and pull requests

#169 - conditioning_gs-xxx.jpg are empty when running main.py

Issue - State: open - Opened by J-1nn about 1 month ago - 4 comments

#154 - Error when running Inversion script

Issue - State: open - Opened by macrocredit over 1 year ago - 7 comments

#100 - Is all the patching in ldm/modules/encoders/modules.py actually necessary?

Issue - State: closed - Opened by zwishenzug about 2 years ago - 2 comments

#100 - Is all the patching in ldm/modules/encoders/modules.py actually necessary?

Issue - State: closed - Opened by zwishenzug about 2 years ago - 2 comments

#99 - Question: How should you read the logged images?

Issue - State: closed - Opened by TaleirOfDeynai about 2 years ago - 4 comments

#99 - Question: How should you read the logged images?

Issue - State: closed - Opened by TaleirOfDeynai about 2 years ago - 4 comments

#98 - >= 1.0 loss

Issue - State: closed - Opened by badcode6 about 2 years ago - 3 comments

#98 - >= 1.0 loss

Issue - State: closed - Opened by badcode6 about 2 years ago - 3 comments

#97 - refactor(load_model_from_config): remove `model.cuda()` call

Pull Request - State: closed - Opened by ryanrussell about 2 years ago

#97 - refactor(load_model_from_config): remove `model.cuda()` call

Pull Request - State: closed - Opened by ryanrussell about 2 years ago

#96 - Best learning rate to immediately overfit asap ?

Issue - State: open - Opened by 1blackbar about 2 years ago - 9 comments

#96 - Best learning rate to immediately overfit asap ?

Issue - State: open - Opened by 1blackbar about 2 years ago - 9 comments

#95 - Adding init image

Pull Request - State: open - Opened by tmuskal about 2 years ago

#95 - Adding init image

Pull Request - State: open - Opened by tmuskal about 2 years ago

#94 - dual 12gb gpu gets CUDA OOM, single works

Issue - State: closed - Opened by ryanrussell about 2 years ago - 9 comments

#94 - dual 12gb gpu gets CUDA OOM, single works

Issue - State: closed - Opened by ryanrussell about 2 years ago - 9 comments

#93 - No `test_dataloader()` method defined to run `Trainer.test`.

Issue - State: closed - Opened by yanzengatsmu about 2 years ago - 1 comment

#93 - No `test_dataloader()` method defined to run `Trainer.test`.

Issue - State: closed - Opened by yanzengatsmu about 2 years ago - 1 comment

#92 - Reproducing the results in the paper

Issue - State: open - Opened by shadow2496 about 2 years ago - 25 comments

#92 - Reproducing the results in the paper

Issue - State: open - Opened by shadow2496 about 2 years ago - 25 comments

#91 - 'the Protocol Buffer runtime library' ERROR

Issue - State: closed - Opened by yanzengatsmu about 2 years ago - 3 comments

#91 - 'the Protocol Buffer runtime library' ERROR

Issue - State: closed - Opened by yanzengatsmu about 2 years ago - 3 comments

#90 - fixing placeholder_strings param

Pull Request - State: closed - Opened by tmuskal about 2 years ago - 1 comment

#90 - fixing placeholder_strings param

Pull Request - State: closed - Opened by tmuskal about 2 years ago - 1 comment

#89 - How about training on more images of a domin? For example, 100~200 images?

Issue - State: closed - Opened by universewill about 2 years ago - 2 comments

#89 - How about training on more images of a domin? For example, 100~200 images?

Issue - State: closed - Opened by universewill about 2 years ago - 2 comments

#88 - What is init_word?

Issue - State: closed - Opened by EKZERSIS about 2 years ago - 3 comments

#88 - What is init_word?

Issue - State: closed - Opened by EKZERSIS about 2 years ago - 3 comments

#87 - UserWarning: Trying to infer the `batch_size` from an ambiguous collection. The batch size we found is 14. To avoid any miscalculations, use `self.log(..., batch_size=batch_size)`.

Issue - State: closed - Opened by EKZERSIS about 2 years ago - 1 comment

#87 - UserWarning: Trying to infer the `batch_size` from an ambiguous collection. The batch size we found is 14. To avoid any miscalculations, use `self.log(..., batch_size=batch_size)`.

Issue - State: closed - Opened by EKZERSIS about 2 years ago - 1 comment

#86 - AttributeError: 'NoneType' object has no attribute 'load'

Issue - State: closed - Opened by Josh-Yang95 about 2 years ago - 7 comments

#86 - AttributeError: 'NoneType' object has no attribute 'load'

Issue - State: closed - Opened by Josh-Yang95 about 2 years ago - 7 comments

#85 - placeholder_tokens and placeholder_strings

Issue - State: closed - Opened by tmuskal about 2 years ago - 1 comment

#85 - placeholder_tokens and placeholder_strings

Issue - State: closed - Opened by tmuskal about 2 years ago - 1 comment

#84 - NameError: name 'trainer' is not defined

Issue - State: closed - Opened by EKZERSIS about 2 years ago - 11 comments

#84 - NameError: name 'trainer' is not defined

Issue - State: closed - Opened by EKZERSIS about 2 years ago - 11 comments

#83 - Artifacts arising from 256x256 data

Issue - State: open - Opened by GucciFlipFlops1917 about 2 years ago - 5 comments

#83 - Artifacts arising from 256x256 data

Issue - State: open - Opened by GucciFlipFlops1917 about 2 years ago - 5 comments

#82 - Reconstruction loss does not change, although samples improving

Issue - State: closed - Opened by RealAntonVoronov about 2 years ago - 2 comments

#82 - Reconstruction loss does not change, although samples improving

Issue - State: closed - Opened by RealAntonVoronov about 2 years ago - 2 comments

#81 - How did you prepare the prompts when training the Sstyle?

Issue - State: closed - Opened by ajundo about 2 years ago - 2 comments

#81 - How did you prepare the prompts when training the Sstyle?

Issue - State: closed - Opened by ajundo about 2 years ago - 2 comments

#80 - AttributeError: module 'signal' has no attribute 'SIGUSR1'

Issue - State: closed - Opened by EKZERSIS about 2 years ago - 3 comments

#80 - AttributeError: module 'signal' has no attribute 'SIGUSR1'

Issue - State: closed - Opened by EKZERSIS about 2 years ago - 3 comments

#79 - Interpret weighted prompts

Pull Request - State: open - Opened by 0x2b3bfa0 about 2 years ago

#79 - Interpret weighted prompts

Pull Request - State: open - Opened by 0x2b3bfa0 about 2 years ago

#78 - AttributeError: 'int' object has no attribute 'strip'

Issue - State: closed - Opened by EKZERSIS about 2 years ago

#77 - Would it be possible to train a multi-vector embedding using masking for features which individual vectors should be perfected for?

Issue - State: closed - Opened by CodeExplode about 2 years ago - 2 comments

#76 - No module named 'taming'

Issue - State: closed - Opened by steven4547466 about 2 years ago - 2 comments

#75 - You have to specify the number of GPUs you would like to use, add `num_processes=...` to your call.

Issue - State: closed - Opened by CptJackieSparrow about 2 years ago - 5 comments

#74 - excuse me,How can I find the embeddings_gs-5049.pt

Issue - State: closed - Opened by mohenghui about 2 years ago - 2 comments

#73 - size mismatch for model.diffusion_model.output_blocks.11.1.transformer_blocks.0.attn2.to_v.weight: copying a param with shape torch.Size([320, 768]) from checkpoint, the shape in current model is torch.Size([320, 1280]).

Issue - State: closed - Opened by mengen-li about 2 years ago - 3 comments

#72 - Process sometimes gets stuck repeatedly running DDIM sampler and creating outputs in images/val (not images/train)

Issue - State: closed - Opened by CodeExplode about 2 years ago - 9 comments

#71 - Model training effect is not good

Issue - State: closed - Opened by Lufffya about 2 years ago - 10 comments

#70 - Input -> Reconstruction fine detail loss shows up in all generated output

Issue - State: closed - Opened by SeverianVoid about 2 years ago - 5 comments

#69 - Multi-gpu?

Issue - State: closed - Opened by dboshardy about 2 years ago - 2 comments

#68 - Outpus by trained model is very different from origin training materials

Issue - State: open - Opened by inspire-boy about 2 years ago - 17 comments

#67 - NameError: name 'trainer' is not defined

Issue - State: closed - Opened by fpena06 about 2 years ago - 7 comments

#66 - Unused validation repeats value in v1-finetune.yaml

Issue - State: closed - Opened by hopibel about 2 years ago - 3 comments

#65 - Allowing initializer words which map to >1 token if num_vectors_per_token supports it

Issue - State: closed - Opened by CodeExplode about 2 years ago - 4 comments

#64 - How to use huggingface concepts Or is there a guide doc of original IMG->pt->SD？

Issue - State: closed - Opened by inspire-boy about 2 years ago - 6 comments

#63 - Suddenly process outputs bad images

Issue - State: closed - Opened by dboshardy about 2 years ago - 4 comments

#62 - Unable to invert - NameError: name 'trainer' is not defined ; UnpicklingError: invalid load key, 'm'.

Issue - State: closed - Opened by EladNoy about 2 years ago - 7 comments

#61 - Cannot get satisfying reconstruction results.

Issue - State: closed - Opened by liuquande about 2 years ago - 16 comments

#60 - Some output images are blurry while others are sharp

Issue - State: closed - Opened by dboshardy about 2 years ago - 4 comments

#59 - Are the 'imagenet_templates_small' templates in personalized.py used just for visual inspection, or actual testing?

Issue - State: closed - Opened by CodeExplode about 2 years ago - 2 comments

#58 - ImportError: cannot import name 'VectorQuantizer2' from 'taming.modules.vqvae.quantize' (/opt/anaconda3/envs/ldm/lib/python3.8/site-packages/taming/modules/vqvae/quantize.py)

Issue - State: closed - Opened by Aditya-Bahl about 2 years ago - 8 comments

#57 - How to generate "a image1[inverse token1] with the style from image2[inverse token2]" with this pretrained model?

Issue - State: closed - Opened by BEaN001 about 2 years ago - 1 comment

#56 - Project page uses artwork of an artist who recently passed away

Issue - State: closed - Opened by mttgll about 2 years ago - 4 comments

#55 - Help with set up please

Issue - State: closed - Opened by Aditya-Bahl about 2 years ago - 2 comments

#54 - 'Upright' images for objects which are often seen from all sorts of rotations - does it matter?

Issue - State: closed - Opened by CodeExplode about 2 years ago - 2 comments

#53 - Possible to find the closest actual word in the dictionary?

Issue - State: closed - Opened by isikdogan about 2 years ago - 2 comments

#52 - How to achieve "Bipartite Inversion" ?

Issue - State: closed - Opened by kingofprank about 2 years ago - 4 comments

#51 - CUDA out of memory. Tried to allocate 2048.00 GiB when it starts running DDIM sampling

Issue - State: closed - Opened by TerraXSoul about 2 years ago - 7 comments

#50 - RuntimeError: CUDA out of memory. Tried to allocate 9.00 GiB (GPU 0; 24.00 GiB total capacity; 4.37 GiB already allocated; 12.90 GiB free; 8.83 GiB reserved in total by PyTorch)

Issue - State: closed - Opened by jamiethemorris about 2 years ago - 18 comments

#49 - Any way to force training into ignoring background and focus just on a subject ?

Issue - State: closed - Opened by 1blackbar about 2 years ago - 14 comments

#48 - Unable to start training

Issue - State: closed - Opened by webmaster94 about 2 years ago - 14 comments

#47 - Per image tokens on a larger dataset

Issue - State: closed - Opened by LexCybermac about 2 years ago - 2 comments

#46 - Inpainting only accepts image and mask

Issue - State: closed - Opened by dboshardy about 2 years ago - 2 comments

#45 - error during training setup: 'int' object has no attribute 'strip'

Issue - State: closed - Opened by nylki about 2 years ago - 2 comments

#44 - 'pytorch_lightning.loggers' has no attribute 'TestTubeLogger'

Issue - State: closed - Opened by TristanMorelle about 2 years ago - 22 comments

#43 - Dtype mismatch when using embeddings with SD

Issue - State: closed - Opened by ClassicDirt about 2 years ago

#42 - size mismatch for model.diffusion_model.input_blocks.1.1.transformer

Issue - State: closed - Opened by TristanMorelle about 2 years ago - 2 comments

#41 - OSError: cannot open resource

Issue - State: closed - Opened by rickyars about 2 years ago - 6 comments

GitHub / rinongal/textual_inversion issues and pull requests