---
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: meta-llama/Meta-Llama-3-8B-Instruct
datasets:
- b-mc2/sql-create-context
model-index:
- name: llama3-8b-instruct-text-to-sql
  results: []
metrics:
- accuracy 79.90
language:
- en
---



# llama3-8b-instruct-text-to-sql

This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the generator dataset.

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 3
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 6
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 3

### Training results



### Framework versions

- PEFT 0.10.0
- Transformers 4.40.0
- Pytorch 2.2.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1


### Train jupyter notebook

[my github](https://github.com/bofen97/llama3-8b-instruct-text-to-sql/blob/main/llama3-8b-instruct-text-to-sql.ipynb)

### Usage

```python

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_id = "tyfeng1997/llama3-8b-instruct-text-to-sql"

tokenizer = AutoTokenizer.from_pretrained(model_id)

model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)


messages = [
    {"role": "system", "content": "You are an text to SQL query translator. Users will ask you questions in English and you will generate a SQL query based on the provided SCHEMA.\nSCHEMA:\nCREATE TABLE match_season (College VARCHAR, POSITION VARCHAR)"},
    {"role": "user", "content": "Which college have both players with position midfielder and players with position defender?"},
]

input_ids = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    return_tensors="pt"
).to(model.device)

terminators = [
    tokenizer.eos_token_id,
    tokenizer.convert_tokens_to_ids("<|eot_id|>")
]

outputs = model.generate(
    input_ids,
    max_new_tokens=256,
    eos_token_id=terminators,
    do_sample=True,
    temperature=0.6,
    top_p=0.9,
)
response = outputs[0]
print(tokenizer.decode(response, skip_special_tokens=True))

#
#system
#You are an text to SQL query translator. Users will ask you questions in English and you will generate a SQL query based on the provided SCHEMA.
#SCHEMA:
#CREATE TABLE match_season (College VARCHAR, POSITION VARCHAR)
#user
#Which college have both players with position midfielder and players with position defender?
#assistant
#SELECT College FROM match_season WHERE POSITION = "Midfielder" INTERSECT SELECT College FROM match_season WHERE POSITION = "Defender"
#
```