---
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
dataset_info:
  features:
  - name: label
    dtype: string
  - name: prompt_similarity
    dtype: float32
  - name: identity_consistency
    dtype: float32
  - name: negative_prompt_similarity
    dtype: float32
  - name: target_prompt_similarity
    dtype: float32
  - name: aesthetic_score
    dtype: float32
  - name: ir_score
    dtype: float64
  splits:
  - name: train
    num_bytes: 466
    num_examples: 14
  download_size: 4388
  dataset_size: 466
---

        method: db_multi_hot_ 

        num_inference_steps: 30

        prompt_similarity : 0.2267446517944336 
identity_consistency : 0.5888814330101013 
negative_prompt_similarity : 0.21684415638446808 
target_prompt_similarity : 0.2199733853340149 
aesthetic_score : 3.8228139877319336 
ir_score : -1.0597369602988833