Downloads · 30 days
15
5% of all-time downloads
gokaygokay/phytia410m_desctoprompt
phytia410m_desctoprompt is a text generation model from gokaygokay. Use it when you need the model to write or continue text. It is set up for transformers.
Fine tuned EleutherAI/pythia-410m using gokaygokay/promptdescriptionstablediffusion3k dataset.
Downloads · 30 days
15
5% of all-time downloads
All-time downloads
285
Public
Parameters
405M
1.6 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors1.6 GB · 100%
From the Hugging Face model README
Fine tuned EleutherAI/pythia-410m using gokaygokay/prompt_description_stable_diffusion_3k dataset.
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "gokaygokay/phytia410m_desctoprompt"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
# Your description
test_description = """
View to a rustic terrace filled with pots with autumn flowers and a vine full of red leaves and bunches of grapes.
in the foreground a wooden table with a copious breakfast, coffee, bowls, vases and plates with fruits, nuts, chestnuts, hazelnuts, breads and buns.
"""
prompt_template = """### Description:
{description}
### Prompt:
"""
text = prompt_template.format(description=test_description)
def inference(text, model, tokenizer, max_input_tokens=1000, max_output_tokens=200):
# Tokenize
input_ids = tokenizer.encode(
text,
return_tensors="pt",
truncation=True,
max_length=max_input_tokens
)
# Generate
device = model.device
generated_tokens_with_prompt = model.generate(
input_ids=input_ids.to(device),
max_length=max_output_tokens,
)
# Decode
generated_text_with_prompt = tokenizer.batch_decode(generated_tokens_with_prompt, skip_special_tokens=True)
# Strip the prompt
generated_text_answer = generated_text_with_prompt[0][len(text):]
return generated_text_answer
print("Description input (test):", text)
print("Finetuned model's prompt: ")
print(inference(text, model, tokenizer))
<!-- gokaygokay-citation -->
This model release is maintained by Gökay Aydoğan. If you reference this repository in academic work, please cite it as follows and also cite the upstream models, datasets, or projects it builds upon.
@software{aydogan2024phytia410m_desctoprompt,
author = {Aydoğan, Gökay},
title = {{phytia410m_desctoprompt}},
year = {2024},
publisher = {Hugging Face},
url = {https://huggingface.co/gokaygokay/phytia410m_desctoprompt},
note = {Model repository; cite the base model and upstream datasets as required.}
}