Model description
This Stable-Diffusion Model has been fine-tuned on images of the Star Trek Voyager Spaceship.
Here are some examples that were created using the model using these settings:
Prompt: photo of voyager spaceship in space, high quality, blender, 3d, trending on artstation, 8k
Negative Prompt: bad, ugly, malformed, deformed, out of frame, blurry
Denoising Steps: 50
Intended uses & limitations
Anyone may use this model for non-commercial usecases under the Linked License, as long as Paragraph 5 of the Open RAIL-M License are respected as well. The original Model adheres under Open RAIL-M.
It was made solely as an experiment for keras_cv Dreambooth Training.
Since a lot of orthographic views were used, the model seems to be biased around them, and has issues creating more variance and poses. While inferring, the background appears noisy.
Training and evaluation data
Images from Rob Bonchune from Trekcore were used for training.
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
Hyperparameters | Value |
---|---|
inner_optimizer.class_name | Custom>RMSprop |
inner_optimizer.config.name | RMSprop |
inner_optimizer.config.weight_decay | None |
inner_optimizer.config.clipnorm | None |
inner_optimizer.config.global_clipnorm | None |
inner_optimizer.config.clipvalue | None |
inner_optimizer.config.use_ema | False |
inner_optimizer.config.ema_momentum | 0.99 |
inner_optimizer.config.ema_overwrite_frequency | 100 |
inner_optimizer.config.jit_compile | True |
inner_optimizer.config.is_legacy_optimizer | False |
inner_optimizer.config.learning_rate | 0.0010000000474974513 |
inner_optimizer.config.rho | 0.9 |
inner_optimizer.config.momentum | 0.0 |
inner_optimizer.config.epsilon | 1e-07 |
inner_optimizer.config.centered | False |
dynamic | True |
initial_scale | 32768.0 |
dynamic_growth_steps | 2000 |
training_precision | mixed_float16 |
- Downloads last month
- 2