flan-t5-large-PLsql / README.md
MRNH's picture
Create README.md
21f47c6
metadata
language:
  - en
pipeline_tag: text2text-generation
metrics:
  - f1
tags:
  - SQL
  - plSQL
  - english

This is a fine-tuned version of T5 FLAN LARGE (783M) on English in particular on the public dataset spider for text-toSQL.

To initialize the model:

from transformers import T5ForConditionalGeneration
model = T5ForConditionalGeneration.from_pretrained("MRNH/flan-t5-large-PLsql")

Use the tokenizer:

tokenizer = T5ForConditionalGeneration.from_pretrained("MRNH/flan-t5-large-PLsql")

input = tokenizer("<question> "+sentence["db_id"]+" </question> "+sentence["question"],
                  text_target=sentence["query"], return_tensors='pt')

To generate text using the model:

output = model.generate(input["input_ids"],attention_mask=input["attention_mask"])

Training of the model is performed using the following loss computation based on the hidden state output h:

h.logits, h.loss = model(input_ids=input["input_ids"],
                                              attention_mask=input["attention_mask"],
                                              labels=input["labels"])