Instructions to use ganchengguang/OIELLM-13B-Instruction with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ganchengguang/OIELLM-13B-Instruction with Transformers:

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

pipe = pipeline("text-generation", model="ganchengguang/OIELLM-13B-Instruction")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ganchengguang/OIELLM-13B-Instruction")
model = AutoModelForCausalLM.from_pretrained("ganchengguang/OIELLM-13B-Instruction")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ganchengguang/OIELLM-13B-Instruction with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ganchengguang/OIELLM-13B-Instruction"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ganchengguang/OIELLM-13B-Instruction",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ganchengguang/OIELLM-13B-Instruction

SGLang

How to use ganchengguang/OIELLM-13B-Instruction 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 "ganchengguang/OIELLM-13B-Instruction" \
    --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": "ganchengguang/OIELLM-13B-Instruction",
		"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 "ganchengguang/OIELLM-13B-Instruction" \
        --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": "ganchengguang/OIELLM-13B-Instruction",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ganchengguang/OIELLM-13B-Instruction with Docker Model Runner:
```
docker model run hf.co/ganchengguang/OIELLM-13B-Instruction
```

This is a model of paper. Base in LLaMA3-8B-Instruction. Meta https://huggingface.co/meta-llama/Meta-Llama-3-8B

Please must use following format to use OIELLM. And extraction information from input text or sentence.

The OIELLM support 3 languages (English, Chinese and Japanese). And you must use task instruct words to define kind of task.

The following is input and output format: { "input": "In 1953, filming of "On the Waterfront" starring Marlon Brando began, and Kazan struggled with Spiegel's persistent budget cuts and managed to complete the film, which was released the following year in 1954 and became a huge hit with support from the laborer class./NER", "output": "Literature/NER/:Person;Marlon Brando:Product Name;On the Waterfront:Person;Kazan:Person;Spiegel" }

The from_pretrain class is use AutoTokenizer and AutoModelForCausalLM.

If you have any question. You can leave the words in this commutiy. Or contact me from paper's E-mail directly.

Let me conclude by thanking the contributors to the MMM dataset for contributing the fundamental dataset. And the pioneering researchers who selflessly contributed.

1. Japanese Wikipedia NER dataset Takahiro Omi https://github.com/stockmarkteam/ner-wikipedia-dataset

2. JGLUE: Japanese General Language Understanding Evaluation Kentaro Kurihara, Daisuke Kawahara, Tomohide Shibata https://github.com/yahoojapan/JGLUE?tab=readme-ov-file

3. livedoor news corpus 関口宏司 https://www.rondhuit.com/download.html

4. UniversalNER Wenxuan Zhou https://arxiv.org/abs/2308.03279

Paper address and cite information: https://arxiv.org/abs/2407.10953

@misc{gan2024mmmmultilingualmutualreinforcement, title={MMM: Multilingual Mutual Reinforcement Effect Mix Datasets & Test with Open-domain Information Extraction Large Language Models}, author={Chengguang Gan and Qingyu Yin and Xinyang He and Hanjun Wei and Yunhao Liang and Younghun Lim and Shijian Wang and Hexiang Huang and Qinghao Zhang and Shiwen Ni and Tatsunori Mori}, year={2024}, eprint={2407.10953}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2407.10953}, }

Downloads last month: 5

Safetensors

Model size

13B params

Tensor type

BF16

Collection including ganchengguang/OIELLM-13B-Instruction

Mutual Reinforcement Effect

Collection

The MRE model and Datasets. • 15 items • Updated Oct 18, 2024 • 1

Papers for ganchengguang/OIELLM-13B-Instruction

MMM: Multilingual Mutual Reinforcement Effect Mix Datasets & Test with Open-domain Information Extraction Large Language Models

Paper • 2407.10953 • Published Jul 15, 2024 • 5

UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognition

Paper • 2308.03279 • Published Aug 7, 2023 • 24