File size: 1,352 Bytes
e322aac
 
 
 
 
 
 
5dc9814
 
 
 
e322aac
 
 
5dc9814
 
e322aac
 
 
5dc9814
e322aac
ab1da75
e322aac
 
 
ab1da75
e322aac
ab1da75
e322aac
ab1da75
 
 
 
 
 
5dc9814
 
ab1da75
 
 
e322aac
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f08e6d3
25124c4
e322aac
d09968e
25124c4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
---
base_model: defog/sqlcoder-7b-2
library_name: peft
license: cc-by-sa-4.0
tags:
- trl
- sft
- QLora
- peft
- SQL
- causal-lm
model-index:
- name: sqlcoder-7b-2_FineTuned_PEFT_QLORA_adapter
  results: []
language:
- en
---


# sqlcoder-7b-2_FineTuned_QLORA_Adapter

This model is a fine-tuned version of [defog/sqlcoder-7b-2](https://huggingface.co/defog/sqlcoder-7b-2) on 260 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

### Framework versions

- PEFT 0.13.2
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.19.1