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

Libraries

How to use numind/NuExtract-large with Transformers:

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

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

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("numind/NuExtract-large", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use numind/NuExtract-large with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "numind/NuExtract-large"
# 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-large",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

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

SGLang

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

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

liamcripwell commited on Jun 25, 2024

Commit

8c54698

verified ·

1 Parent(s): 0b2054b

Update README.md

Browse files

Files changed (1) hide show

README.md +5 -5

README.md CHANGED Viewed

@@ -5,7 +5,7 @@ language:
 ---
 # Structure Extraction Model by NuMind 🔥
-NuExtract-large is a fine-tuned version of [phi-3-small](https://huggingface.co/microsoft/Phi-3-small-8k-instruct), on a private high-quality syntatic dataset for information extraction.
 To use the model, provide an input text (less than 2000 tokens) and a JSON schema describing the information you need to extract.
 Note: This model is purely extractive, so each information output by the model is present as it is in the text. You can also provide an example of output to help the model understand your task more precisely.
@@ -45,7 +45,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
-def predict_NuExtract(model,tokenizer,text, schema,example = ["","",""]):
     schema = json.dumps(json.loads(schema), indent=4)
     input_llm =  "<|input|>\n### Template:\n" +  schema + "\n"
     for i in example:
@@ -53,13 +53,13 @@ def predict_NuExtract(model,tokenizer,text, schema,example = ["","",""]):
           input_llm += "### Example:\n"+ json.dumps(json.loads(i), indent=4)+"\n"
     input_llm +=  "### Text:\n"+text +"\n<|output|>\n"
-    input_ids = tokenizer(input_llm, return_tensors="pt",truncation = True, max_length = 4000).to("cuda")
     output = tokenizer.decode(model.generate(**input_ids)[0], skip_special_tokens=True)
     return output.split("<|output|>")[1].split("<|end-output|>")[0]
-model = AutoModelForCausalLM.from_pretrained("numind/NuExtract", trust_remote_code=True,torch_dtype=torch.bfloat16)
 tokenizer = AutoTokenizer.from_pretrained("numind/NuExtract", trust_remote_code=True)
 model.to("cuda")
@@ -91,7 +91,7 @@ schema = """{
     }
 }"""
-prediction = predict_NuExtract(model,tokenizer,text, schema,example = ["","",""])
 print(prediction)
 ```

 ---
 # Structure Extraction Model by NuMind 🔥
+NuExtract-large is a fine-tuned version of [phi-3-small](https://huggingface.co/microsoft/Phi-3-small-8k-instruct), on a private high-quality synthetic dataset for information extraction.
 To use the model, provide an input text (less than 2000 tokens) and a JSON schema describing the information you need to extract.
 Note: This model is purely extractive, so each information output by the model is present as it is in the text. You can also provide an example of output to help the model understand your task more precisely.
 import torch
+def predict_NuExtract(model, tokenizer, text, schema, example=["","",""]):
     schema = json.dumps(json.loads(schema), indent=4)
     input_llm =  "<|input|>\n### Template:\n" +  schema + "\n"
     for i in example:
           input_llm += "### Example:\n"+ json.dumps(json.loads(i), indent=4)+"\n"
     input_llm +=  "### Text:\n"+text +"\n<|output|>\n"
+    input_ids = tokenizer(input_llm, return_tensors="pt", truncation=True, max_length=4000).to("cuda")
     output = tokenizer.decode(model.generate(**input_ids)[0], skip_special_tokens=True)
     return output.split("<|output|>")[1].split("<|end-output|>")[0]
+model = AutoModelForCausalLM.from_pretrained("numind/NuExtract", trust_remote_code=True, torch_dtype=torch.bfloat16)
 tokenizer = AutoTokenizer.from_pretrained("numind/NuExtract", trust_remote_code=True)
 model.to("cuda")
     }
 }"""
+prediction = predict_NuExtract(model, tokenizer, text, schema, example=["","",""])
 print(prediction)
 ```