Instructions to use piazzola/address-detection-model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use piazzola/address-detection-model with Transformers:

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

pipe = pipeline("text-generation", model="piazzola/address-detection-model")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("piazzola/address-detection-model")
model = AutoModelForCausalLM.from_pretrained("piazzola/address-detection-model", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use piazzola/address-detection-model with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/piazzola/address-detection-model

SGLang

How to use piazzola/address-detection-model 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 "piazzola/address-detection-model" \
    --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": "piazzola/address-detection-model",
		"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 "piazzola/address-detection-model" \
        --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": "piazzola/address-detection-model",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use piazzola/address-detection-model with Docker Model Runner:
```
docker model run hf.co/piazzola/address-detection-model
```

tmp_trainer

This model is a fine-tuned version of facebook/opt-350m on the addressWithContext dataset.

Model description

Make sure to set max_new_tokens = 20; otherwise, the model will generate one token at a time.

nlp = pipeline("text-generation",
                model="piazzola/tmp_trainer",
                max_new_tokens=20)
                
nlp("I live at 15 Firstfield Road.")

Note that if you would like to try longer sentences using the Hosted inference API on the right hand side on this website, you might need to click "Compute" more than one time to get the address.

Intended uses & limitations

The model is intended to detect addresses that occur in a sentence.

Training and evaluation data

This model is trained on piazzola/addressWithContext.

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3.0

Framework versions

Transformers 4.34.0
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.14.1

Downloads last month: 34

Model tree for piazzola/address-detection-model

Base model

facebook/opt-350m

Finetuned

(161)

this model