Home

Project outline

1. In progress

1.A. Optimizing the minimal UI/UX through Colab forms 
1.B. Improving code annotation
1.C. Improving code structure / optimizing code
1.D. Setting up the storage / collection module to store the pictures in a way that trades memory space for GPU calculation

2. Next steps to setup

2.A. Implementing a prompt generator linked to GPT-J in the workflow
2.B. Creating chaos within the prompt through random insertion of characters to add another layer of variability
2.C. Offering more than one choice of upscale
2.D. Optimizing the workflow so it runs only inside one cell of the colab
2.E. Autosaving each production within google drive
2.F. Setting UP a Flask / Gradio UI than runs whithin colab to allow access to the GPU

3. Completed Steps

3.A. Implementing ESRGAN in the workflow
3.B. Setting the seed and settings collection workflow for the three step process
3.C. Working-out the variation process and setting it up in the workflow
3.D. Adding randomness within the first generation processing including size/proportions (whitin a computable range), number of inference steps     (q/quality) and guidance (g/guid) values.