Preface
In the past I was not happy with locomotives I created. I used the steam locomotive to test some kohya_ss parameters. Therefore this model is not optimal, but it works.
Trigger Words
The trigger words for this LoRA are
steam locomotiveTraining Images
The training image have an arbitrary resolution, an arbitrary aspect ration and an arbirtray orientation.
I used in total 29 arbitrary images. I have chosen the photos absolutely at random.
Training settings was
29 Images
145 buckets
5 repeats
Batch size 3
220 epochs (-> used 90)
2860 learning steps (-> used 2175 training steps)
Quality of The Created Images
My first tests shows that the quality of the created images are depending on the model. Some models are better with respect to a steam locomotive than others.
What the Model Does
Creates a photo realistic steam locomotive.
Weights
You can use weights between 0.0 and 0.6. Over the weight of 0.6 overfitting will occur and distorted images are the result.
Donation
About each Buzz I am happy! Yellow ones are very welcome!
I am also happy to accept donations in one of 250+ crypto currencies:
https://nowpayments.io/donation/zentrocdot
Leave optional a comment like improve documentation or improve model etc. when doing so, that I know in what you are interested in.