| Property | Value |
|---|---|
| Base Model | Stable Diffusion XL 1.0 |
| License | Apache 2.0 |
| Network Dimensions | 64 |
| Network Alpha | 32 |
What is Canopus-Textile-Pattern-adp-LoRA?
Canopus-Textile-Pattern-adp-LoRA is a specialized LoRA adaptation for Stable Diffusion XL that focuses on generating high-quality textile patterns. The model has been trained specifically to produce detailed 4K resolution outputs of various textile patterns, including traditional Kalamkari, Ikat, and floral designs.
Implementation Details
The model utilizes AdamW optimizer with a constant learning rate scheduler and incorporates advanced noise processing techniques. Training was conducted over 9 epochs with 1080 total steps, using specific parameters like noise offset (0.03) and multires noise iterations (10) for optimal pattern generation.
- Network Architecture: 64 dimensions with 32 alpha
- Training Configuration: 30 repeats with epoch saves
- Specialized noise processing with 0.1 discount rate
Core Capabilities
- High-resolution 4K textile pattern generation
- Specialized in Kalamkari, Ikat, and floral patterns
- Seamless integration with SDXL base model
- Optimized for detail preservation in pattern designs
Frequently Asked Questions
Q: What makes this model unique?
This model specializes in generating highly detailed textile patterns with specific expertise in traditional designs like Kalamkari and Ikat, utilizing optimized parameters for pattern consistency and detail preservation.
Q: What are the recommended use cases?
The model is ideal for designers and artists looking to generate authentic textile patterns, particularly for digital textile design, fashion visualization, and interior design applications.