Data

Tools

Indexes

Applications

Experiments

Open Source Science

242-continuous-subject-specific-attribute-control-in-t2i-models-by-identifying-semantic-directions

https://github.com/szu-advtech-2024/242-continuous-subject-specific-attribute-control-in-t2i-models-by-identifying-semantic-directions

Science Score: 13.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ○
    .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (3.4%) to scientific vocabulary
Last synced: 10 months ago · JSON representation

Repository

Basic Info
  • Host: GitHub
  • Owner: SZU-AdvTech-2024
  • Default Branch: main
  • Size: 0 Bytes
Statistics
  • Stars: 0
  • Watchers: 0
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 1 year ago · Last pushed over 1 year ago
Metadata Files
Citation

https://github.com/SZU-AdvTech-2024/242-Continuous-Subject-Specific-Attribute-Control-in-T2I-Models-by-Identifying-Semantic-Directions/blob/main/ 

# Continuous, Subject-Specific Attribute Control in T2I Models by Identifying Semantic Directions

## 

,(Text to Image,T2I).,(""),.,,.,CLIP,.,,.,(),,.

## 

Githubhttps://github.com/CompVis/attribute-control

READMEREADME



## 



- python3.9.20
- CUDA 12.4

requirements.txt



```bash
pip install -r requirements.txt
```

## 

stable-diffusion-xl-base-1.0

## 



```bash
python learn_delta.py device=cuda:0 model=sdxl prompts=people/age
```



```bash
python learn_delta.py device=cuda:0 model=sdxl prompts=people/age model.compile=False +model.gradient_checkpointing=True
```

## 

notebooks/inference_sdxl.ipynbblock3

```python
state_dict = torch.load('enter your delta result here')
#
```

## 



- pretrained_deltas

- outputs

  - outputs\learn_delta\people\age\runs\2024-11-24\02-28-23

    - 

      ```python
      #
      max_steps: 1000
      batch_size: 1
      grad_accum_steps: 8
      scale_batch_size: 4
      ```

  - outputs\learn_delta\people\age\runs\2024-12-06\14-25-43

    - 

      ```python
      #
      max_steps: 2000
      batch_size: 1
      grad_accum_steps: 8
      scale_batch_size: 4
      ```

  - outputs\learn_delta\people\age\runs\2025-01-07\07-35-18

    - 

      ```python
      #
      max_steps: 1000
      batch_size: 1
      grad_accum_steps: 5
      scale_batch_size: 4
      ```

  - outputs\learn_delta\people\age\runs\2025-01-08\08-57-38

    - 

      ```python
      #
      max_steps: 500
      batch_size: 1
      grad_accum_steps: 8
      scale_batch_size: 4
      ```

Owner

  • Name: SZU-AdvTech-2024
  • Login: SZU-AdvTech-2024
  • Kind: organization

GitHub Events

Total
  • Push event: 2
  • Create event: 3
Last Year
  • Push event: 2
  • Create event: 3