|
--- |
|
license: creativeml-openrail-m |
|
tags: |
|
- stable-diffusion |
|
- prompt-generator |
|
widget: |
|
- text: "amazing" |
|
- text: "a photo of" |
|
- text: "a sci-fi" |
|
- text: "a portrait of" |
|
- text: "a person standing" |
|
- text: "a boy watching" |
|
datasets: |
|
- poloclub/diffusiondb |
|
- Gustavosta/Stable-Diffusion-Prompts |
|
- bartman081523/stable-diffusion-discord-prompts |
|
- FredZhang7/krea-ai-prompts |
|
--- |
|
# DistilGPT2 Stable Diffusion V2 Model Card |
|
This model was trained on 2.47 million descriptive stable diffusion prompts on the [FredZhang7/distilgpt2-stable-diffusion](https://huggingface.co/FredZhang7/distilgpt2-stable-diffusion) checkpoint for another 4.27 million steps. |
|
|
|
Compared to other prompt generation models using GPT2, this one runs with 50% faster forwardpropagation and 40% less disk space & RAM. |
|
|
|
Major improvements from v1 are: |
|
- 25% more variations |
|
- more capable of generating story-like prompts |
|
- cleaned training data |
|
* removed prompts that generate images with nsfw scores > 0.5 |
|
* removed duplicates, including prompts that differ by capitalization and punctuations |
|
* removed punctuations at random places |
|
* removed prompts shorter than 15 characters |
|
|
|
|
|
### PyTorch |
|
|
|
```bash |
|
pip install --upgrade transformers |
|
``` |
|
|
|
Faster but less fluent generation: |
|
|
|
```python |
|
from transformers import GPT2Tokenizer, GPT2LMHeadModel |
|
tokenizer = GPT2Tokenizer.from_pretrained('distilgpt2') |
|
tokenizer.add_special_tokens({'pad_token': '[PAD]'}) |
|
model = GPT2LMHeadModel.from_pretrained('FredZhang7/distilgpt2-stable-diffusion-v2', pad_token_id=tokenizer.eos_token_id) |
|
|
|
prompt = r'a cat sitting' |
|
|
|
# generate text using fine-tuned model |
|
from transformers import pipeline |
|
nlp = pipeline('text-generation', model=model, tokenizer=tokenizer) |
|
|
|
# generate 5 samples |
|
outs = nlp(prompt, max_length=80, num_return_sequences=5) |
|
|
|
print('\nInput:\n' + 100 * '-') |
|
print('\033[96m' + prompt + '\033[0m') |
|
print('\nOutput:\n' + 100 * '-') |
|
for i in range(len(outs)): |
|
outs[i] = str(outs[i]['generated_text']).replace(' ', '') |
|
print('\033[92m' + '\n\n'.join(outs) + '\033[0m\n') |
|
``` |
|
|
|
Example output: |
|
![greedy search](./greedy_search.png) |
|
|
|
<br> |
|
|
|
Slower but more fluent generation: |
|
|
|
```python |
|
from transformers import GPT2Tokenizer, GPT2LMHeadModel |
|
tokenizer = GPT2Tokenizer.from_pretrained('distilgpt2') |
|
tokenizer.add_special_tokens({'pad_token': '[PAD]'}) |
|
model = GPT2LMHeadModel.from_pretrained('FredZhang7/distilgpt2-stable-diffusion-v2') |
|
model.eval() |
|
|
|
prompt = r'a cat sitting' # the beginning of the prompt |
|
temperature = 0.9 # a higher temperature will produce more diverse results, but with a higher risk of less coherent text. |
|
top_k = 8 # the number of tokens to sample from at each step |
|
max_length = 80 # the maximum number of tokens for the output of the model |
|
repitition_penalty = 1.2 # the penalty value for each repetition of a token |
|
num_return_sequences=5 # the number of results to generate |
|
|
|
# generate the result with contrastive search. generate 5 results with the highest probability out of 10. |
|
input_ids = tokenizer(prompt, return_tensors='pt').input_ids |
|
output = model.generate(input_ids, do_sample=True, temperature=temperature, top_k=top_k, max_length=max_length, num_return_sequences=num_return_sequences, repetition_penalty=repitition_penalty, penalty_alpha=0.6, no_repeat_ngram_size=1, early_stopping=True) |
|
|
|
print('\nInput:\n' + 100 * '-') |
|
print('\033[96m' + prompt + '\033[0m') |
|
print('\nOutput:\n' + 100 * '-') |
|
for i in range(len(output)): |
|
print('\033[92m' + tokenizer.decode(output[i], skip_special_tokens=True) + '\033[0m\n') |
|
``` |
|
|
|
Example output: |
|
![constrastive search](./constrastive_search.png) |