Modelpublic

keras-dreambooth/pink-floyd-division-bell

sourceHugging Facecreativeml-openrail-mupdated 4y agoView on Hugging Face

1likes7downloads

Model Card

Model description

This model is a stable diffusion keras model fine tuned using dreambooth. This model has been fine tuned to learn the concept of pink floyd's division bell album cover. To generate images including this concept, use this in your prompt - "pink floyd division bell album cover".

Intended uses & limitations

It can be used to generate images of your choice using a text prompt.

Training data

pink floyd division bell album cover 0

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

Hyperparameters	Value
inneroptimizer.classname	Custom>RMSprop
inner_optimizer.config.name	RMSprop
inneroptimizer.config.weightdecay	None
inner_optimizer.config.clipnorm	None
inneroptimizer.config.globalclipnorm	None
inner_optimizer.config.clipvalue	None
inneroptimizer.config.useema	False
inneroptimizer.config.emamomentum	0.99
inneroptimizer.config.emaoverwrite_frequency	100
inneroptimizer.config.jitcompile	True
inneroptimizer.config.islegacy_optimizer	False
inneroptimizer.config.learningrate	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
dynamicgrowthsteps	2000
training_precision	mixed_float16

## Model Plot

<details> <summary>View Model Plot</summary>

[image]

</details>

This model was contributed by Shivalika Singh