Instructions to use yuanzhoulvpi/vit-gpt2-image-chinese-captioning with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yuanzhoulvpi/vit-gpt2-image-chinese-captioning with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="yuanzhoulvpi/vit-gpt2-image-chinese-captioning")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("yuanzhoulvpi/vit-gpt2-image-chinese-captioning")
model = AutoModelForMultimodalLM.from_pretrained("yuanzhoulvpi/vit-gpt2-image-chinese-captioning", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yuanzhoulvpi/vit-gpt2-image-chinese-captioning with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yuanzhoulvpi/vit-gpt2-image-chinese-captioning"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yuanzhoulvpi/vit-gpt2-image-chinese-captioning",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/yuanzhoulvpi/vit-gpt2-image-chinese-captioning

SGLang

How to use yuanzhoulvpi/vit-gpt2-image-chinese-captioning with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "yuanzhoulvpi/vit-gpt2-image-chinese-captioning" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yuanzhoulvpi/vit-gpt2-image-chinese-captioning",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "yuanzhoulvpi/vit-gpt2-image-chinese-captioning" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yuanzhoulvpi/vit-gpt2-image-chinese-captioning",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use yuanzhoulvpi/vit-gpt2-image-chinese-captioning with Docker Model Runner:
```
docker model run hf.co/yuanzhoulvpi/vit-gpt2-image-chinese-captioning
```

模型介绍

vit对图像做encoder，然后再用gpt2做decoder
vit模型使用的是google/vit-base-patch16-224, gpt2使用的是yuanzhoulvpi/gpt2_chinese
本模型支持中文

训练代码

https://github.com/yuanzhoulvpi2017/zero_nlp/tree/main/vit-gpt2-image-chinese-captioning

推理代码

infer

from transformers import (VisionEncoderDecoderModel, 
                          AutoTokenizer,ViTImageProcessor)
import torch
from PIL import Image

vision_encoder_decoder_model_name_or_path = "yuanzhoulvpi/vit-gpt2-image-chinese-captioning"#"vit-gpt2-image-chinese-captioning/checkpoint-3200"

processor = ViTImageProcessor.from_pretrained(vision_encoder_decoder_model_name_or_path)
tokenizer = AutoTokenizer.from_pretrained(vision_encoder_decoder_model_name_or_path)
model = VisionEncoderDecoderModel.from_pretrained(vision_encoder_decoder_model_name_or_path)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)

max_length = 16
num_beams = 4
gen_kwargs = {"max_length": max_length, "num_beams": num_beams}


def predict_step(image_paths):
    images = []
    for image_path in image_paths:
        i_image = Image.open(image_path)
        if i_image.mode != "RGB":
            i_image = i_image.convert(mode="RGB")

        images.append(i_image)

    pixel_values = processor(images=images, return_tensors="pt").pixel_values
    pixel_values = pixel_values.to(device)

    output_ids = model.generate(pixel_values, **gen_kwargs)

    preds = tokenizer.batch_decode(output_ids, skip_special_tokens=True)
    preds = [pred.strip() for pred in preds]
    return preds


predict_step(['bigdata/image_data/train-1000200.jpg'])

效果

example 1

example 2

Downloads last month: 59

Safetensors

Model size

0.2B params

Tensor type

F32