meQ_model

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

• Loss: 1.1869
• Rouge1: 0.4942
• Rouge2: 0.346
• Rougel: 0.4743
• Rougelsum: 0.474
• Gen Len: 13.61

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: 2e-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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.835	1.0	57	2.0475	0.2345	0.0895	0.2074	0.2067	16.82
2.3806	2.0	114	1.7243	0.2854	0.126	0.2649	0.2667	15.83
2.1701	3.0	171	1.5659	0.3633	0.211	0.3445	0.3444	14.72
2.0194	4.0	228	1.4690	0.4301	0.2752	0.4115	0.4127	13.95
1.9218	5.0	285	1.4159	0.468	0.3125	0.4499	0.4493	13.26
1.8645	6.0	342	1.3811	0.487	0.3462	0.472	0.4715	13.22
1.8254	7.0	399	1.3494	0.4836	0.3424	0.4668	0.466	13.17
1.7843	8.0	456	1.3277	0.4818	0.3444	0.4676	0.4672	13.04
1.7543	9.0	513	1.3084	0.4775	0.3412	0.463	0.4627	13.15
1.7206	10.0	570	1.2961	0.476	0.3393	0.4602	0.4617	13.13
1.6977	11.0	627	1.2823	0.4762	0.3395	0.4603	0.462	13.18
1.6725	12.0	684	1.2701	0.4841	0.3437	0.4677	0.4694	13.28
1.6479	13.0	741	1.2649	0.4912	0.3505	0.4755	0.4778	13.37
1.6313	14.0	798	1.2546	0.4896	0.344	0.4724	0.4742	13.47
1.6154	15.0	855	1.2488	0.4898	0.3456	0.4738	0.476	13.48
1.5932	16.0	912	1.2433	0.4935	0.3506	0.4776	0.4806	13.47
1.5716	17.0	969	1.2347	0.4984	0.3529	0.4789	0.4815	13.46
1.5523	18.0	1026	1.2314	0.4881	0.3456	0.4713	0.4722	13.48
1.5393	19.0	1083	1.2277	0.4925	0.35	0.4754	0.4761	13.57
1.535	20.0	1140	1.2239	0.4866	0.3415	0.4693	0.4708	13.63
1.5389	21.0	1197	1.2178	0.4785	0.3359	0.463	0.4621	13.56
1.5203	22.0	1254	1.2132	0.4837	0.3362	0.4679	0.4682	13.75
1.4909	23.0	1311	1.2098	0.4877	0.3393	0.4716	0.4719	13.71
1.4957	24.0	1368	1.2102	0.4874	0.3393	0.4713	0.4714	13.66
1.4746	25.0	1425	1.2076	0.4881	0.3398	0.4725	0.4717	13.66
1.4745	26.0	1482	1.2041	0.496	0.3474	0.4799	0.4792	13.64
1.4605	27.0	1539	1.2040	0.4903	0.3416	0.4741	0.4733	13.65
1.4465	28.0	1596	1.2024	0.4961	0.3461	0.4793	0.4784	13.7
1.4398	29.0	1653	1.2006	0.4859	0.3385	0.4692	0.4698	13.65
1.4469	30.0	1710	1.1976	0.4887	0.3426	0.473	0.4718	13.69
1.4218	31.0	1767	1.1965	0.4934	0.3469	0.4778	0.4764	13.64
1.4315	32.0	1824	1.1966	0.488	0.3447	0.4726	0.472	13.51
1.4282	33.0	1881	1.1957	0.488	0.3447	0.4726	0.472	13.51
1.396	34.0	1938	1.1932	0.489	0.3459	0.4739	0.4729	13.52
1.4028	35.0	1995	1.1941	0.4892	0.3434	0.4723	0.4722	13.63
1.4068	36.0	2052	1.1922	0.4895	0.347	0.4733	0.4722	13.57
1.3831	37.0	2109	1.1911	0.4927	0.3451	0.4742	0.474	13.63
1.3781	38.0	2166	1.1903	0.4896	0.3434	0.4717	0.4714	13.57
1.3867	39.0	2223	1.1889	0.4915	0.3464	0.4736	0.4729	13.63
1.3694	40.0	2280	1.1893	0.492	0.3444	0.4728	0.4723	13.58
1.3912	41.0	2337	1.1891	0.4902	0.3448	0.4719	0.4713	13.46
1.3793	42.0	2394	1.1886	0.492	0.3444	0.4728	0.4723	13.58
1.3664	43.0	2451	1.1884	0.4907	0.3434	0.4717	0.4714	13.53
1.3787	44.0	2508	1.1874	0.4919	0.3442	0.4725	0.472	13.61
1.3692	45.0	2565	1.1871	0.4919	0.3442	0.4725	0.472	13.61
1.3732	46.0	2622	1.1875	0.492	0.3444	0.4728	0.4723	13.58
1.3752	47.0	2679	1.1872	0.4942	0.346	0.4743	0.474	13.61
1.3581	48.0	2736	1.1871	0.4942	0.346	0.4743	0.474	13.61
1.3509	49.0	2793	1.1869	0.4942	0.346	0.4743	0.474	13.61
1.3752	50.0	2850	1.1869	0.4942	0.346	0.4743	0.474	13.61

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.1+cu116
• Datasets 2.11.0
• Tokenizers 0.13.2