Instructions to use numind/NuExtract-tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use numind/NuExtract-tiny with Transformers:

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

pipe = pipeline("text-generation", model="numind/NuExtract-tiny")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("numind/NuExtract-tiny")
model = AutoModelForCausalLM.from_pretrained("numind/NuExtract-tiny")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use numind/NuExtract-tiny with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/numind/NuExtract-tiny

SGLang

How to use numind/NuExtract-tiny 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 "numind/NuExtract-tiny" \
    --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": "numind/NuExtract-tiny",
		"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 "numind/NuExtract-tiny" \
        --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": "numind/NuExtract-tiny",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use numind/NuExtract-tiny with Docker Model Runner:
```
docker model run hf.co/numind/NuExtract-tiny
```

Fine Tuning

by skylerjoe - opened Jul 5, 2024

Discussion

skylerjoe

Jul 5, 2024

•

edited Jul 5, 2024

Hi everyone, thanks for putting this great model out there. I was fine-tuning the larger 3.8B parameter version using PEFT and creating an adaptor using AutoTrain. However, the sizes of the model and the adaptor created using AutoTrain are different.
''' my error
size mismatch for base_model.model.model.embed_tokens.weight: copying a param with shape torch.Size([32012, 3072]) from checkpoint, the shape in current model is torch.Size([32064, 3072]).
size mismatch for base_model.model.lm_head.weight: copying a param with shape torch.Size([32012, 3072])
from checkpoint, the shape in current model is torch.Size([32064, 3072]).
'''
It seems as though it has something to do with the vocabulary size. If anyone could help me or point me to some resources that would be great! Thanks in advance.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment