metadata

license: mit
datasets:
  - b-mc2/sql-create-context
  - gretelai/synthetic_text_to_sql
language:
  - en

Fine-Tuned Google T5 Model for Text to SQL Translation

A fine-tuned version of the Google T5 model, trained for the task of translating natural language queries into SQL statements.

Model Details

Architecture: Google T5 Base (Text-to-Text Transfer Transformer)
Task: Text to SQL Translation
Fine-Tuning Datasets:
- sql-create-context Dataset
- Synthetic-Text-To-SQL Dataset

Ongoing Work

Currently working to implement PICARD (Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models) to improve the results of this model. More details can be found in the original PICARD paper.

Results

Results are currently being evaluated and will be posted here soon.

Files

optimizer.pt: State of the optimizer.
training_args.bin: Training arguments and hyperparameters.
tokenizer.json: Tokenizer vocabulary and settings.
spiece.model: SentencePiece model file.
special_tokens_map.json: Special tokens mapping.
tokenizer_config.json: Tokenizer configuration settings.
model.safetensors: Trained model weights.
generation_config.json: Configuration for text generation.
config.json: Model architecture configuration.