kijai/comfyui-fluxtrainer issues and pull requests

#44 - How to continue training the previous model?

Issue - State: open - Opened by chaorenai 2 months ago - 7 comments

#44 - How to continue training the previous model?

Issue - State: open - Opened by chaorenai 2 months ago - 7 comments

#43 - import failed

Issue - State: open - Opened by groktheo 2 months ago - 1 comment

#43 - import failed

Issue - State: open - Opened by groktheo 2 months ago - 1 comment

#42 - step number when restarting

Issue - State: open - Opened by chrisgoringe 2 months ago

#42 - step number when restarting

Issue - State: open - Opened by chrisgoringe 2 months ago

#41 - processing interrupted

Issue - State: closed - Opened by RaySteve312 2 months ago - 2 comments

#41 - processing interrupted

Issue - State: closed - Opened by RaySteve312 2 months ago - 2 comments

#40 - after I trained 1000 steps, I want to change the learning rate but not start from the 0step.

Issue - State: closed - Opened by RaySteve312 2 months ago - 1 comment

#40 - after I trained 1000 steps, I want to change the learning rate but not start from the 0step.

Issue - State: closed - Opened by RaySteve312 2 months ago - 1 comment

#39 - HOW to train concrete blocks?

Issue - State: closed - Opened by Maelstrom2014 2 months ago - 4 comments

#39 - HOW to train concrete blocks?

Issue - State: closed - Opened by Maelstrom2014 2 months ago - 4 comments

#38 - Exploding the pot, it is determined that the folder does not exist. Please create it first

Issue - State: open - Opened by StartHua 2 months ago - 1 comment

#37 - feature request - reg data set

Issue - State: closed - Opened by whmc76 2 months ago - 5 comments

#37 - feature request - reg data set

Issue - State: closed - Opened by whmc76 2 months ago - 5 comments

#36 - Black Image After Training and No Training Loss / Moving Tensors From Meta Device Error

Issue - State: closed - Opened by corbin-hayden13 2 months ago - 2 comments

#36 - Black Image After Training and No Training Loss / Moving Tensors From Meta Device Error

Issue - State: closed - Opened by corbin-hayden13 2 months ago - 2 comments

#35 - save and resume trainining

Issue - State: closed - Opened by kotaxyz 2 months ago - 11 comments

#35 - save and resume trainining

Issue - State: closed - Opened by kotaxyz 2 months ago - 11 comments

#34 - Can i input the parameter "gradient accumulation steps" in some way?

Issue - State: open - Opened by RaySteve312 2 months ago - 8 comments

#34 - Can i input the parameter "gradient accumulation steps" in some way?

Issue - State: open - Opened by RaySteve312 2 months ago - 8 comments

#33 - error while running the nodes

Issue - State: closed - Opened by kotaxyz 3 months ago - 8 comments

#33 - error while running the nodes

Issue - State: closed - Opened by kotaxyz 3 months ago - 8 comments

#32 - I don't know why the LoRA for training failed

Issue - State: open - Opened by chenminglin 3 months ago - 5 comments

#32 - I don't know why the LoRA for training failed

Issue - State: open - Opened by chenminglin 3 months ago - 5 comments

#31 - workflow json file to use lora trained with this method?

Issue - State: open - Opened by summerstay 3 months ago

#31 - workflow json file to use lora trained with this method?

Issue - State: open - Opened by summerstay 3 months ago

#30 - adafactor scheduler error IndexError: list index out of range

Issue - State: open - Opened by whmc76 3 months ago - 3 comments

#30 - adafactor scheduler error IndexError: list index out of range

Issue - State: open - Opened by whmc76 3 months ago - 3 comments

#29 - Out of Memory for Init Training

Issue - State: open - Opened by markleaf131313 3 months ago - 1 comment

#29 - Out of Memory for Init Training

Issue - State: open - Opened by markleaf131313 3 months ago - 1 comment

#28 - "enable fp8 training for U-Net" how to disable it?

Issue - State: closed - Opened by RaySteve312 3 months ago - 4 comments

#28 - "enable fp8 training for U-Net" how to disable it?

Issue - State: closed - Opened by RaySteve312 3 months ago - 4 comments

#27 - No module named 'networks.lora_flux'

Issue - State: open - Opened by Chzzll 3 months ago - 4 comments

#27 - No module named 'networks.lora_flux'

Issue - State: open - Opened by Chzzll 3 months ago - 4 comments

#26 - T5 Encoder in unet training

Issue - State: closed - Opened by bananasss00 3 months ago - 1 comment

#26 - T5 Encoder in unet training

Issue - State: closed - Opened by bananasss00 3 months ago - 1 comment

#25 - InitFluxLoRATraining Error: not enough values to unpack (expected 3, got 2)

Issue - State: open - Opened by EuphoricPenguin 3 months ago - 21 comments

#25 - InitFluxLoRATraining Error: not enough values to unpack (expected 3, got 2)

Issue - State: open - Opened by EuphoricPenguin 3 months ago - 21 comments

#24 - How to train more optimization steps?

Issue - State: closed - Opened by enkie358 3 months ago - 2 comments

#24 - How to train more optimization steps?

Issue - State: closed - Opened by enkie358 3 months ago - 2 comments

#23 - Error occurred when executing InitFluxLoRATraining: extra keys not allowed @ data['general']['alpha_mask']

Issue - State: open - Opened by rwywenyi 3 months ago - 1 comment

#23 - Error occurred when executing InitFluxLoRATraining: extra keys not allowed @ data['general']['alpha_mask']

Issue - State: open - Opened by rwywenyi 3 months ago - 1 comment

#22 - If you have issues loading the dataset images and text...

Issue - State: open - Opened by tonynoce 3 months ago - 1 comment

#22 - If you have issues loading the dataset images and text...

Issue - State: open - Opened by tonynoce 3 months ago - 1 comment

#21 - Error after ComfyUI update: InitFluxLoRATraining Error

Issue - State: open - Opened by Orenji-Tangerine 3 months ago - 3 comments

#21 - Error after ComfyUI update: InitFluxLoRATraining Error

Issue - State: open - Opened by Orenji-Tangerine 3 months ago - 3 comments

#20 - torch.OutOfMemoryError: Allocation on device

Issue - State: open - Opened by funwithforks 3 months ago - 3 comments

#19 - 'FluxNetworkTrainer' object has no attribute 'num_train_epochs'

Issue - State: open - Opened by Luckyjjjjjjj 3 months ago - 11 comments

#19 - 'FluxNetworkTrainer' object has no attribute 'num_train_epochs'

Issue - State: open - Opened by Luckyjjjjjjj 3 months ago - 11 comments

#18 - Can't load tokenizer for 'google/t5-v1_1-xxl'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'google/t5-v1_1-xxl' is the correct path to a directory containing all relevant files for a T5TokenizerFast tokenizer.

Issue - State: open - Opened by rwywenyi 3 months ago - 3 comments

#18 - Can't load tokenizer for 'google/t5-v1_1-xxl'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'google/t5-v1_1-xxl' is the correct path to a directory containing all relevant files for a T5TokenizerFast tokenizer.

Issue - State: open - Opened by rwywenyi 3 months ago - 3 comments

#17 - Resume Training

Issue - State: open - Opened by davi-min 3 months ago - 11 comments

#16 - TypeError: unsupported operand type(s) for +: 'NoneType' and 'float'

Issue - State: closed - Opened by bananasss00 3 months ago - 1 comment

#15 - Prodigy optimizer and weight_decay

Issue - State: closed - Opened by bananasss00 3 months ago - 2 comments

#15 - Prodigy optimizer and weight_decay

Issue - State: closed - Opened by bananasss00 3 months ago - 2 comments

#14 - 'utf-8' codec can't decode byte 0xa1 in position 681: invalid start byte

Issue - State: open - Opened by pandayummy 3 months ago - 1 comment

#14 - 'utf-8' codec can't decode byte 0xa1 in position 681: invalid start byte

Issue - State: open - Opened by pandayummy 3 months ago - 1 comment

#13 - Lora Extract node creates non-functional lora in ComfyUI

Issue - State: open - Opened by TripleHeadedMonkey 3 months ago - 5 comments

#13 - Lora Extract node creates non-functional lora in ComfyUI

Issue - State: open - Opened by TripleHeadedMonkey 3 months ago - 5 comments

#12 - Inference tensors cannot be saved for backward. To work around you can make a clone to get a normal tensor and use it in autograd.

Issue - State: open - Opened by Orenji-Tangerine 3 months ago - 9 comments

#12 - Inference tensors cannot be saved for backward. To work around you can make a clone to get a normal tensor and use it in autograd.

Issue - State: open - Opened by Orenji-Tangerine 3 months ago - 9 comments

#11 - Train lora with image width more than 1024 px, like 960*1280

Issue - State: closed - Opened by Lantianyou 3 months ago - 6 comments

#11 - Train lora with image width more than 1024 px, like 960*1280

Issue - State: closed - Opened by Lantianyou 3 months ago - 6 comments

#10 - AttributeError: 'FluxNetworkTrainer' object has no attribute 'num_train_epochs'

Issue - State: closed - Opened by whmc76 3 months ago - 1 comment

#10 - AttributeError: 'FluxNetworkTrainer' object has no attribute 'num_train_epochs'

Issue - State: closed - Opened by whmc76 3 months ago - 1 comment

#9 - ERROR !!! Exception during processing !!! Reserved escape sequence used (line execution.py:3867 column 1 char 122)

Issue - State: closed - Opened by whmc76 3 months ago

#8 - Various missing nodes - suggest update to Installation instructions

Issue - State: closed - Opened by ianmcintyre 3 months ago - 1 comment

#7 - Finished training cant find lora

Issue - State: closed - Opened by Reefermann94 3 months ago - 2 comments

#6 - Train fail. Error occurred when executing InitFluxLoRATraining: Cannot copy out of meta tensor; no data!

Issue - State: closed - Opened by ultimatech-cn 3 months ago - 9 comments

#5 - Unable to load vocabulary from file.

Issue - State: open - Opened by Celtmant 3 months ago - 1 comment

#4 - Add pyproject.toml for Custom Node Registry

Pull Request - State: closed - Opened by robinjhuang 3 months ago

#3 - Add Github Action for Publishing to Comfy Registry

Pull Request - State: closed - Opened by robinjhuang 3 months ago

#2 - ImageConcatFromBatch node not found

Issue - State: closed - Opened by ShmuelRonen 3 months ago - 1 comment

#1 - You Absolute Legend`

Issue - State: open - Opened by MushroomFleet 3 months ago

GitHub / kijai/comfyui-fluxtrainer issues and pull requests