flan-t5-small-20-epochs-fine-tune

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

• Loss: 1.1136
• Accuracy: 1
• F1 Micro: 1
• F1 Macro: 1
• F1 Weighted: 1

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Micro	F1 Macro	F1 Weighted
2.0193	1.0	643	1.3910	1	1	1	1
1.3609	2.0	1286	0.9353	1	1	1	1
1.0198	3.0	1929	0.8408	1	1	1	1
0.7264	4.0	2572	0.7945	1	1	1	1
0.6595	5.0	3215	0.7847	1	1	1	1
0.5689	6.0	3858	0.7898	1	1	1	1
0.4745	7.0	4501	0.7969	1	1	1	1
0.4204	8.0	5144	0.8213	1	1	1	1
0.403	9.0	5787	0.8726	1	1	1	1
0.3578	10.0	6430	0.8777	1	1	1	1
0.3238	11.0	7073	0.9143	1	1	1	1
0.287	12.0	7716	0.9656	1	1	1	1
0.2903	13.0	8359	0.9580	1	1	1	1
0.2463	14.0	9002	1.0306	1	1	1	1
0.2318	15.0	9645	1.0428	1	1	1	1
0.2265	16.0	10288	1.0483	1	1	1	1
0.1954	17.0	10931	1.0825	1	1	1	1
0.191	18.0	11574	1.0972	1	1	1	1
0.1774	19.0	12217	1.1163	1	1	1	1
0.1847	20.0	12860	1.1136	1	1	1	1

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.2
• Datasets 2.1.0
• Tokenizers 0.15.2