README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: t5-small-asqa-ob
    results: []

t5-small-asqa-ob

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

• Loss: 2.9381
• Rouge1: 0.1633
• Rouge2: 0.0907
• Rougel: 0.1394
• Rougelsum: 0.1393

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: 0.0005
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
3.8212	1.0	710	2.7920	0.1248	0.0624	0.1064	0.1063
3.0559	2.0	1420	2.5937	0.1319	0.0715	0.1139	0.1138
2.568	3.0	2130	2.4971	0.1398	0.0754	0.1206	0.1204
2.384	4.0	2840	2.5024	0.1473	0.0817	0.1273	0.1271
2.1599	5.0	3550	2.4947	0.1498	0.0824	0.1288	0.1287
2.0444	6.0	4260	2.5305	0.1502	0.0837	0.1291	0.1290
1.9219	7.0	4970	2.5486	0.1599	0.0890	0.1376	0.1373
1.7532	8.0	5680	2.5772	0.1647	0.0914	0.1413	0.1411
1.6895	9.0	6390	2.6346	0.1630	0.0911	0.1397	0.1395
1.5751	10.0	7100	2.6650	0.1700	0.0944	0.1450	0.1449
1.4616	11.0	7810	2.6705	0.1571	0.0874	0.1348	0.1346
1.3923	12.0	8520	2.7767	0.1695	0.0951	0.1453	0.1450
1.3043	13.0	9230	2.8091	0.1704	0.0943	0.1460	0.1457
1.2868	14.0	9940	2.8390	0.1553	0.0854	0.1327	0.1324
1.176	15.0	10650	2.9381	0.1633	0.0907	0.1394	0.1393

Framework versions

• Transformers 4.23.0.dev0
• Pytorch 1.12.1+cu102
• Datasets 2.5.1
• Tokenizers 0.12.1