File size: 1,732 Bytes
486892c f83505d 486892c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 |
---
datasets:
- AnyaSchen/image2poetry_ru
language:
- ru
tags:
- Mayakovsky
- image2poetry
---
This repo contains model for generation poetry in style of Mayakovsky from image. The model is fune-tuned concatecation of two pre-trained models: [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) as encoder and [AnyaSchen/rugpt3_mayak](https://huggingface.co/AnyaSchen/rugpt3_mayak) as decoder.
To use this model you can do:
```
from PIL import Image
import requests
from transformers import AutoTokenizer, VisionEncoderDecoderModel, ViTImageProcessor
def generate_poetry(fine_tuned_model, image, tokenizer):
pixel_values = feature_extractor(images=image, return_tensors="pt").pixel_values
pixel_values = pixel_values.to(device)
# Generate the poetry with the fine-tuned VisionEncoderDecoder model
generated_tokens = fine_tuned_model.generate(
pixel_values,
max_length=300,
num_beams=3,
top_p=0.8,
temperature=2.0,
do_sample=True,
pad_token_id=tokenizer.pad_token_id,
eos_token_id=tokenizer.eos_token_id,
)
# Decode the generated tokens
generated_poetry = tokenizer.decode(generated_tokens[0], skip_special_tokens=True)
return generated_poetry
path = 'AnyaSchen/vit-rugpt3-medium-mayak'
fine_tuned_model = VisionEncoderDecoderModel.from_pretrained(path).to(device)
feature_extractor = ViTImageProcessor.from_pretrained(path)
tokenizer = AutoTokenizer.from_pretrained(path)
url = 'https://anandaindia.org/wp-content/uploads/2018/12/happy-man.jpg'
image = Image.open(requests.get(url, stream=True).raw)
generated_poetry = generate_poetry(fine_tuned_model, image, tokenizer)
print(generated_poetry)
``` |