metadata
license: apache-2.0
datasets:
- MohamedRashad/midjourney-detailed-prompts
language:
- en
base_model:
- stable-diffusion-v1-5/stable-diffusion-v1-5
pipeline_tag: text-to-image
library_name: diffusers
tags:
- art
Training Configuration
Base model: runwayml/stable-diffusion-v1-5
Dataset: /kaggle/input/midjourney-v0/midjourney-512px-v0
LoRA rank: 62
SNR gamma: 5.0
Seed: 42
| Parameter | Value |
|---|---|
| Train steps | 1000 |
| Epochs | 11 |
| Steps per epoch | 94 |
| Batch size | 4 |
| Accumulation steps | 2 |
| Effective batch size | 8 |
| Learning rate | 8e-05 |
| Max grad norm | 1.0 |