SQL_Tuning_runpod

This model is a fine-tuned version of google/flan-t5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0452
• Bleu: 44.1094
• Gen Len: 18.9135

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: 0.0001
• 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: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.2746	0.08	1000	0.1082	42.9089	18.9275
0.1472	0.15	2000	0.0852	43.3232	18.9021
0.1233	0.23	3000	0.0736	43.5297	18.915
0.1121	0.31	4000	0.0662	43.6957	18.9164
0.1	0.39	5000	0.0611	43.8882	18.9131
0.092	0.46	6000	0.0576	43.9239	18.9136
0.0874	0.54	7000	0.0531	43.8973	18.9138
0.0824	0.62	8000	0.0515	43.9284	18.911
0.0788	0.69	9000	0.0495	43.9772	18.9131
0.0765	0.77	10000	0.0472	44.057	18.9152
0.0723	0.85	11000	0.0463	44.0788	18.915
0.0708	0.93	12000	0.0452	44.1094	18.9135

Framework versions

• Transformers 4.33.3
• Pytorch 2.1.0.dev20230322+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3