flant5action

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: 0.1428
• Rouge1: 56.0664
• Rouge2: 34.7343
• Rougel: 56.0394
• Rougelsum: 56.0313
• Gen Len: 18.9852

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.2525	1.0	674	0.2294	53.2181	29.8509	53.1635	53.1474	19.0
0.2434	2.0	1348	0.2240	53.5453	30.1367	53.4479	53.44	18.9970
0.2281	3.0	2022	0.2135	53.1901	30.3456	53.0849	53.0759	18.9970
0.2221	4.0	2696	0.2056	52.0669	29.4321	51.9567	51.9512	18.9881
0.2145	5.0	3370	0.2012	54.484	31.6451	54.4213	54.4144	18.9970
0.2121	6.0	4044	0.1961	54.1219	31.2019	54.0701	54.0668	18.9970
0.1979	7.0	4718	0.1901	54.9091	32.2416	54.8482	54.8318	18.9911
0.2086	8.0	5392	0.1846	54.9615	32.4701	54.8836	54.8821	18.9970
0.1985	9.0	6066	0.1795	55.2027	32.5792	55.1531	55.1431	18.9970
0.2027	10.0	6740	0.1746	54.4079	32.2598	54.38	54.3697	18.9970
0.1922	11.0	7414	0.1707	55.4814	33.2069	55.4428	55.4298	18.9970
0.1806	12.0	8088	0.1660	55.7189	33.831	55.6796	55.6702	18.9970
0.1834	13.0	8762	0.1623	55.6253	33.9516	55.5925	55.585	18.9941
0.1795	14.0	9436	0.1596	55.6786	33.7589	55.6232	55.6183	18.9911
0.1767	15.0	10110	0.1553	55.8132	34.1603	55.795	55.7873	18.9911
0.1792	16.0	10784	0.1539	55.9694	34.4612	55.9454	55.9323	18.9792
0.1785	17.0	11458	0.1521	56.2202	34.6224	56.1781	56.1706	18.9941
0.1705	18.0	12132	0.1496	56.4102	34.7821	56.3911	56.3789	18.9911
0.1668	19.0	12806	0.1478	56.1222	34.6804	56.0821	56.077	18.9881
0.1729	20.0	13480	0.1459	56.1605	34.8596	56.1349	56.1221	18.9852
0.1759	21.0	14154	0.1451	56.1232	34.8956	56.1054	56.0994	18.9852
0.1713	22.0	14828	0.1439	55.9801	34.6435	55.9556	55.9482	18.9763
0.1751	23.0	15502	0.1436	56.2088	34.8754	56.1771	56.1758	18.9852
0.1626	24.0	16176	0.1431	56.0657	34.7302	56.04	56.0317	18.9852
0.1696	25.0	16850	0.1428	56.0664	34.7343	56.0394	56.0313	18.9852

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.3
• Tokenizers 0.13.3