sqlcoder-7b-2_FineTuned_PEFT_QLORA_adapter_alpha_r_32

This model is a fine-tuned version of defog/sqlcoder-7b-2 on 260 MS SQL examples (Task, Schema and Answer triplets) related to financial/banking domain.

Intended uses & limitations

MS SQL Server - SQL Query Generation

Training

This model was trained using the QLoRA method with the following configurations:

• r = 64,
• lora_alpha = 32
• lora_dropout = 0.05
• bias='none'
• task_type='CAUSAL_LM'

Quantization parameters:

• load_in_4bit=True
• bnb_4bit_quant_type="nf4"
• bnb_4bit_compute_dtype=torch.bfloat16

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Framework versions

• PEFT 0.13.1
• Transformers 4.44.2
• Pytorch 2.3.1+cu121
• Datasets 3.0.2
• Tokenizers 0.19.1