Instructions to use cfahlgren1/NaturalSQL-6.7B-v0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cfahlgren1/NaturalSQL-6.7B-v0 with Transformers:

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

pipe = pipeline("text-generation", model="cfahlgren1/NaturalSQL-6.7B-v0")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cfahlgren1/NaturalSQL-6.7B-v0")
model = AutoModelForCausalLM.from_pretrained("cfahlgren1/NaturalSQL-6.7B-v0", device_map="auto")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cfahlgren1/NaturalSQL-6.7B-v0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cfahlgren1/NaturalSQL-6.7B-v0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cfahlgren1/NaturalSQL-6.7B-v0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cfahlgren1/NaturalSQL-6.7B-v0

SGLang

How to use cfahlgren1/NaturalSQL-6.7B-v0 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 "cfahlgren1/NaturalSQL-6.7B-v0" \
    --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": "cfahlgren1/NaturalSQL-6.7B-v0",
		"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 "cfahlgren1/NaturalSQL-6.7B-v0" \
        --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": "cfahlgren1/NaturalSQL-6.7B-v0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cfahlgren1/NaturalSQL-6.7B-v0 with Docker Model Runner:
```
docker model run hf.co/cfahlgren1/NaturalSQL-6.7B-v0
```

Inquiry on Resolving Vocab Size Discrepancy Between Tokenizer and Model

by SEONGRYEONG - opened Feb 13, 2024

Discussion

SEONGRYEONG

Feb 13, 2024

Hello,

I am currently utilizing a Transformer-based model you developed and have encountered an issue where there is a discrepancy in vocab_size between the tokenizer and the model. Specifically, the vocab_size used by the tokenizer is smaller than what is utilized by the model. This mismatch is hindering my ability to effectively leverage the model.

Could you provide any recommendations or possible approaches to resolve this issue? I am particularly interested in methods for aligning the tokenizer's vocab_size with that of the model, or alternatively, reducing the model's vocab_size to match the tokenizer. Additionally, I would appreciate insights into the causes of such discrepancies and measures to prevent them in the future.

Your expertise and advice on this matter would be invaluable. I would also be grateful for any guidance on key considerations to take into account when addressing this issue.

Thank you for your time and assistance.

Best regards,

SeongRyeong

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