metadata

license: apache-2.0
tags:
  - generated_from_trainer
widget:
  - text: >-
      translate to SQL: How many models with BERT architecture are in the
      HuggingFace Hub?
  - text: >-
      translate to English: SELECT COUNT Model FROM table WHERE Architecture =
      RoBERTa AND creator = Manuel Romero
metrics:
  - bleu
model-index:
  - name: t5-small-finetuned-wikisql-sql-nl-nl-sql
    results: []

t5-small-finetuned-wikisql-sql-nl-nl-sql

This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.1932
Bleu: 41.8787
Gen Len: 16.6251

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: 5e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.2655	1.0	8097	0.2252	39.7999	16.6893
0.2401	2.0	16194	0.2066	40.9456	16.6712
0.2236	3.0	24291	0.1985	41.3509	16.5884
0.2158	4.0	32388	0.1944	41.6988	16.6165
0.2122	5.0	40485	0.1932	41.8787	16.6251

Framework versions

Transformers 4.18.0
Pytorch 1.10.0+cu111
Datasets 2.0.0
Tokenizers 0.11.6