|
--- |
|
license: apache-2.0 |
|
library_name: peft |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
base_model: mistralai/Mistral-7B-Instruct-v0.1 |
|
model-index: |
|
- name: Mistral-7B-text-to-sql |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# Mistral-7B-text-to-sql |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) on the generator dataset. |
|
https://huggingface.co/datasets/b-mc2/sql-create-context |
|
|
|
b-mc2/sql-create-context |
|
|
|
## USE CASE |
|
|
|
import torch |
|
from peft import AutoPeftModelForCausalLM |
|
from transformers import AutoTokenizer, pipeline |
|
|
|
peft_model_id = "frankmorales2020/Mistral-7B-text-to-sql" |
|
|
|
|
|
# Load Model with PEFT adapter |
|
|
|
model = AutoPeftModelForCausalLM.from_pretrained( |
|
peft_model_id, |
|
device_map="auto", |
|
torch_dtype=torch.float16 |
|
) |
|
|
|
tokenizer = AutoTokenizer.from_pretrained(peft_model_id) |
|
|
|
# Load into the pipeline |
|
|
|
pipe = pipeline("text-generation", model=model, tokenizer=tokenizer) |
|
|
|
|
|
|
|
## DATASET |
|
|
|
https://huggingface.co/datasets/b-mc2/sql-create-context |
|
|
|
## ARTICLE |
|
https://medium.com/@frankmorales_91352/text-to-sql-generation-a-comprehensive-overview-6feb24f69f3c |
|
|
|
|
|
## 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 |
|
|
|
When evaluated on 1000 samples from the evaluation dataset, our model achieved an impressive accuracy of 76.30%. However, there's room for improvement. We could enhance the model's performance by exploring techniques like few-shot learning, RAG, and Self-healing to generate the SQL query. |
|
|
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.9.0 |
|
- Transformers 4.38.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
