Instructions to use tdoehmen/starcoder-schemapile-fk with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tdoehmen/starcoder-schemapile-fk with Transformers:

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

pipe = pipeline("text-generation", model="tdoehmen/starcoder-schemapile-fk")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tdoehmen/starcoder-schemapile-fk")
model = AutoModelForCausalLM.from_pretrained("tdoehmen/starcoder-schemapile-fk")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use tdoehmen/starcoder-schemapile-fk with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/tdoehmen/starcoder-schemapile-fk

SGLang

How to use tdoehmen/starcoder-schemapile-fk 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 "tdoehmen/starcoder-schemapile-fk" \
    --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": "tdoehmen/starcoder-schemapile-fk",
		"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 "tdoehmen/starcoder-schemapile-fk" \
        --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": "tdoehmen/starcoder-schemapile-fk",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use tdoehmen/starcoder-schemapile-fk with Docker Model Runner:
```
docker model run hf.co/tdoehmen/starcoder-schemapile-fk
```

SchemaPile Foreign Key Detection Model (Starcoder)

Model Description

In this repository we are introducing starcoder-schemapile-fk. It's a language model, based on BigCode/starcoder fine-tuned for predicting foreign key relationships in relational database schemas.

Training Data

Forein key pairs extracted from SchemaPile-Perm, a large collection of relational database schemas.

Evaluation Data

We evaluate the foreign key detection accuracy of starcoder-schemapile-fk and t5-schemapile-fk on schemas from Spider, BIRD-SQL, and CTU PRLR.

Training Procedure

The model was trained, using 4x A100 40GB GPUs with DeepSpeed ZeRO-3 offloading, and following hyperparamters:

learning_rate: 2.0e-05
num_train_epochs: 3
gradient_accumulation_steps: 8
per_device_train_batch_size: 4
bf16: true
warmup_ratio: 0.03
weight_decay: 0.0

See Training Code.

How to Use

We recommend using the provided prompt template and constrained output using jsonformer:

Example Prompt:

You are given the following SQL database tables: 
staff(staff_id, staff_address_id, nickname, first_name, middle_name, last_name, date_of_birth, date_joined_staff, date_left_staff)
addresses(address_id, line_1_number_building, city, zip_postcode, state_province_county, country)
Output a json string with the following schema {table, column, referencedTable, referencedColumn} that contains the foreign key relationship between the two tables.

Example Output:

{'table': 'staff',
 'column': 'staff_address_id',
 'referencedTable': 'addresses',
 'referencedColumn': 'address_id'}

To run the model locally, we recommend using our end-to-end Example Notebook (requires a single A100 40GB).

Downloads last month: 3

Paper for tdoehmen/starcoder-schemapile-fk

The CTU Prague Relational Learning Repository

Paper • 1511.03086 • Published Nov 10, 2015