Everything is WOOD
Description
FAQ
Comments (14)
This looks fantastic!
Would you mind sharing your training data?
You can upload it to the model by clicking Manage Model etc.
Here's a guide: https://civitai.com/articles/2138/lora-training-data-upload-to-civitai
I'm so sorry. I train this model using 113 (still on my PC) this morning. I follow this guide (https://civitai.com/models/22530/guide-make-your-own-loras-easy-and-free) but I deleted output files except the one I've uploaded here on Civitai. Maybe I can replay the entire process.
Ah, what a shame the source images you used to train it with are gone.
It would be fun to see what kind of data you used.
I always keep mine around and try to share them if someone else wants to have a look.
I think he means the input images you trained with? I was considering doing this too, but for figured woods like curly, flamed and quilted maple, various burls, etc. I'm having to search for new images as my old ones are flat textures maps. Great for seeing the figuring and 3D work... but I'm guessing it'd be better to train on objects made from these woods too? Tables, sculpture, etc. So it would be interesting to see what you used.
@mnemic I think your quest is awesome. There's too little good info on training and input images ... particularly other than just ripping some anime char.
*using 113 images .. . My bad. I still have 113 images used to train the LoRA (75 Mb). I used hires images of wooden toys i.e. trucks, cars, helicopters, planes. If you like I can share (somehow) this images.
I added a zip file containing all used images
@Jentix Amazing! Thank you. Really interesting to see how you made a material morph like this. It's pretty good to have lots of real images, compared to ice cream sculptures of things :D
@Jentix Ah, that's fantastic! I don't know if my previous comment made it (looks like it's gone). But thanks for sharing!
I linked to your model in my article on training data so people can find it and learn from it! I learned stuff from it already :D
@mnemic it's thanks to your comment.
Hello,I want to ask you about the pretrained-model you used,I woula be very grateful if you could share ,I am a beginner,trying to train
I used some photos. I used https://colab.research.google.com/github/hollowstrawberry/kohya-colab/blob/main/Dataset_Maker.ipynb#scrollTo=HuJB7BGAyZCw
to create the LoRA dataset.
After creation of dataset I used Lora_Trainer.ipynb - Colaboratory
to effectively train my LoRA
I used 113 images .. . https://civitai.com/api/download/models/213746?type=Training%20Data
My LoRA is here .. . https://civitai.com/api/download/models/213746?type=Model&format=SafeTensor
@Jentix thank you!!
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.













