Edit model card

code-llama-7b-text-to-sql

This model is a fine-tuned version of codellama/CodeLlama-7b-hf 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 0.7.2.dev0
  • Transformers 4.36.2
  • Pytorch 2.1.2
  • Datasets 2.16.1
  • Tokenizers 0.15.2
Downloads last month
4
Inference API
Unable to determine this model’s pipeline type. Check the docs .

Model tree for czkaiwebBusiness/code-llama-7b-text-to-sql

Adapter
(233)
this model