Instructions to use Lin-Chen/ShareCaptioner-Video with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Lin-Chen/ShareCaptioner-Video with Transformers:

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

pipe = pipeline("text-generation", model="Lin-Chen/ShareCaptioner-Video", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Lin-Chen/ShareCaptioner-Video", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Lin-Chen/ShareCaptioner-Video with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Lin-Chen/ShareCaptioner-Video"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lin-Chen/ShareCaptioner-Video",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Lin-Chen/ShareCaptioner-Video

SGLang

How to use Lin-Chen/ShareCaptioner-Video 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 "Lin-Chen/ShareCaptioner-Video" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lin-Chen/ShareCaptioner-Video",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "Lin-Chen/ShareCaptioner-Video" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lin-Chen/ShareCaptioner-Video",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Lin-Chen/ShareCaptioner-Video with Docker Model Runner:
```
docker model run hf.co/Lin-Chen/ShareCaptioner-Video
```

ShareCaptioner-Video / ixc_utils.py

chenlin

init

7f57424 almost 2 years ago

raw

history blame contribute delete

1.12 kB

	import torch
	import numpy as np
	import torchvision
	from PIL import Image
	from torchvision.transforms.functional import InterpolationMode
	import torchvision.transforms as transforms

	def padding_336(b):
	width, height = b.size
	tar = int(np.ceil(height / 336) * 336)
	top_padding = int((tar - height)/2)
	bottom_padding = tar - height - top_padding
	left_padding = 0
	right_padding = 0
	b = transforms.functional.pad(b, [left_padding, top_padding, right_padding, bottom_padding], fill=[255,255,255])

	return b

	def HD_transform(img, hd_num=16):
	width, height = img.size
	trans = False
	if width < height:
	img = img.transpose(Image.TRANSPOSE)
	trans = True
	width, height = img.size
	ratio = (width/ height)
	scale = 1
	while scale*np.ceil(scale/ratio) <= hd_num:
	scale += 1
	scale -= 1
	new_w = int(scale * 336)
	new_h = int(new_w / ratio)

	img = transforms.functional.resize(img, [new_h, new_w],)
	img = padding_336(img)
	width, height = img.size
	if trans:
	img = img.transpose(Image.TRANSPOSE)

	return img