Full Training Tutorial and Guide and Research For a FLUX Style
•
3
Furkan Gözükara
MonsterMMORPG
AI & ML interests
Check out my youtube page SECourses for Stable Diffusion tutorials. They will help you tremendously in every topic
Articles
Organizations
Posts
35
Post
19
Single Block / Layer FLUX LoRA Training Research Results and LoRA Network Alpha Change Impact With LoRA Network Rank Dimension
Full article posted here : https://medium.com/@furkangozukara/single-block-layer-flux-lora-training-research-results-and-lora-network-alpha-change-impact-with-e713cc89c567
Conclusions
As expected, as you train lesse parameters e.g. LoRA vs Full Fine Tuning or Single Blocks LoRA vs all Blocks LoRA, your quality get reduced
Of course you earn some extra VRAM memory reduction and also some reduced size on the disk
Moreover, lesser parameters reduces the overfitting and realism of the FLUX model, so if you are into stylized outputs like comic, it may work better
Furthermore, when you reduce LoRA Network Rank, keep original Network Alpha unless you are going to do a new Learning Rate research
Finally, very best and least overfitting is achieved with full Fine Tuning
Check figure 3 and figure 4 last columns — I make extracted LoRA Strength / Weight 1.1 instead of 1.0
Full fine tuning configs and instructions > https://www.patreon.com/posts/112099700
Second best one is extracting a LoRA from Fine Tuned model if you need a LoRA
Check figure 3 and figure 4 last columns — I make extracted LoRA Strength / Weight 1.1 instead of 1.0
Extract LoRA guide (public article) : https://www.patreon.com/posts/112335162
Third is doing a all layers regular LoRA training
Full guide, configs and instructions > https://www.patreon.com/posts/110879657
And the worst quality is training lesser blocks / layers with LoRA
Full configs are included in > https://www.patreon.com/posts/110879657
So how much VRAM and Speed single block LoRA training brings?
All layers 16 bit is 27700 MB (4.85 second / it) and 1 single block is 25800 MB (3.7 second / it)
All layers 8 bit is 17250 MB (4.85 second / it) and 1 single block is 15700 MB (3.8 second / it)
Image Raw Links
Figure 0 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Full article posted here : https://medium.com/@furkangozukara/single-block-layer-flux-lora-training-research-results-and-lora-network-alpha-change-impact-with-e713cc89c567
Conclusions
As expected, as you train lesse parameters e.g. LoRA vs Full Fine Tuning or Single Blocks LoRA vs all Blocks LoRA, your quality get reduced
Of course you earn some extra VRAM memory reduction and also some reduced size on the disk
Moreover, lesser parameters reduces the overfitting and realism of the FLUX model, so if you are into stylized outputs like comic, it may work better
Furthermore, when you reduce LoRA Network Rank, keep original Network Alpha unless you are going to do a new Learning Rate research
Finally, very best and least overfitting is achieved with full Fine Tuning
Check figure 3 and figure 4 last columns — I make extracted LoRA Strength / Weight 1.1 instead of 1.0
Full fine tuning configs and instructions > https://www.patreon.com/posts/112099700
Second best one is extracting a LoRA from Fine Tuned model if you need a LoRA
Check figure 3 and figure 4 last columns — I make extracted LoRA Strength / Weight 1.1 instead of 1.0
Extract LoRA guide (public article) : https://www.patreon.com/posts/112335162
Third is doing a all layers regular LoRA training
Full guide, configs and instructions > https://www.patreon.com/posts/110879657
And the worst quality is training lesser blocks / layers with LoRA
Full configs are included in > https://www.patreon.com/posts/110879657
So how much VRAM and Speed single block LoRA training brings?
All layers 16 bit is 27700 MB (4.85 second / it) and 1 single block is 25800 MB (3.7 second / it)
All layers 8 bit is 17250 MB (4.85 second / it) and 1 single block is 15700 MB (3.8 second / it)
Image Raw Links
Figure 0 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Post
889
How to Extract LoRA from FLUX Fine Tuning / DreamBooth Training Full Tutorial and Comparison Between Fine Tuning vs Extraction vs LoRA Training
Full article is here public post : https://www.patreon.com/posts/112335162
This was short on length so check out the full article - public post
Conclusions as below
Conclusions
With same training dataset (15 images used), same number of steps (all compared trainings are 150 epoch thus 2250 steps), almost same training duration, Fine Tuning / DreamBooth training of FLUX yields the very best results
So yes Fine Tuning is the much better than LoRA training itself
Amazing resemblance, quality with least amount of overfitting issue
Moreover, extracting a LoRA from Fine Tuned full checkpoint, yields way better results from LoRA training itself
Extracting LoRA from full trained checkpoints were yielding way better results in SD 1.5 and SDXL as well
Comparison of these 3 is made in Image 5 (check very top of the images to see)
640 Network Dimension (Rank) FP16 LoRA takes 6.1 GB disk space
You can also try 128 Network Dimension (Rank) FP16 and different LoRA strengths during inference to make it closer to Fine Tuned model
Moreover, you can try Resize LoRA feature of Kohya GUI but hopefully it will be my another research and article later
Image Raw Links
Image 1 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Image 2 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Image 3 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Image 4 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Image 5 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Full article is here public post : https://www.patreon.com/posts/112335162
This was short on length so check out the full article - public post
Conclusions as below
Conclusions
With same training dataset (15 images used), same number of steps (all compared trainings are 150 epoch thus 2250 steps), almost same training duration, Fine Tuning / DreamBooth training of FLUX yields the very best results
So yes Fine Tuning is the much better than LoRA training itself
Amazing resemblance, quality with least amount of overfitting issue
Moreover, extracting a LoRA from Fine Tuned full checkpoint, yields way better results from LoRA training itself
Extracting LoRA from full trained checkpoints were yielding way better results in SD 1.5 and SDXL as well
Comparison of these 3 is made in Image 5 (check very top of the images to see)
640 Network Dimension (Rank) FP16 LoRA takes 6.1 GB disk space
You can also try 128 Network Dimension (Rank) FP16 and different LoRA strengths during inference to make it closer to Fine Tuned model
Moreover, you can try Resize LoRA feature of Kohya GUI but hopefully it will be my another research and article later
Image Raw Links
Image 1 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Image 2 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Image 3 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Image 4 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Image 5 : MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
models
58
MonsterMMORPG/FLUX-Fine-Tuning-Grid-Tests
Updated
MonsterMMORPG/Stable-Diffusion
Updated
•
4
MonsterMMORPG/mich
Updated
MonsterMMORPG/YoloModels
Updated
MonsterMMORPG/Best_FLUX_LoRA_Checkpoints
Updated
MonsterMMORPG/256_Imgs_FLUX_LoRA_Trainings_Best
Updated
MonsterMMORPG/OneTrainer_FLUX_Training
Updated
MonsterMMORPG/256_images_tests
Updated
MonsterMMORPG/SECourses
Updated
•
38
MonsterMMORPG/FLUX_Kohya_SS_Massive_Research_Part6_T5_XXL
Updated