t5-vietnamese-summarization

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

• Loss: 4.6288
• Rouge1: 0.4728
• Rouge2: 0.1669
• Rougel: 0.3049
• Rougelsum: 0.3049
• Gen Len: 18.7458

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: 70

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
5.2487	1.0	2007	5.0028	0.4671	0.1595	0.2994	0.2994	18.7618
5.217	2.0	4014	4.9802	0.4639	0.1569	0.2984	0.2983	18.7747
5.2191	3.0	6021	4.9685	0.4644	0.1594	0.2989	0.2989	18.7613
5.2254	4.0	8028	4.9477	0.4648	0.1586	0.2988	0.2987	18.7458
5.1735	5.0	10035	4.9366	0.4654	0.1593	0.2988	0.2987	18.761
5.1735	6.0	12042	4.9214	0.4676	0.1611	0.3004	0.3004	18.78
5.1653	7.0	14049	4.9095	0.4681	0.1616	0.3007	0.3007	18.7523
5.1154	8.0	16056	4.8971	0.4664	0.1598	0.3002	0.3001	18.7655
5.1232	9.0	18063	4.8882	0.4683	0.1612	0.3008	0.3008	18.761
5.0995	10.0	20070	4.8758	0.4709	0.1618	0.3021	0.302	18.7518
5.1012	11.0	22077	4.8689	0.4685	0.1616	0.3011	0.3009	18.7665
5.0916	12.0	24084	4.8486	0.4695	0.1623	0.3024	0.3023	18.7655
5.0559	13.0	26091	4.8409	0.4699	0.1631	0.3024	0.3023	18.7849
5.0633	14.0	28098	4.8326	0.4705	0.1613	0.302	0.302	18.7583
5.0335	15.0	30105	4.8243	0.4696	0.1612	0.3023	0.3022	18.7638
5.0271	16.0	32112	4.8046	0.4691	0.1618	0.3022	0.3022	18.7518
5.0045	17.0	34119	4.8060	0.4708	0.1629	0.3029	0.3028	18.7568
5.0072	18.0	36126	4.7945	0.4702	0.1633	0.3024	0.3023	18.776
4.9954	19.0	38133	4.7894	0.47	0.1639	0.3022	0.3021	18.7785
4.9994	20.0	40140	4.7773	0.4692	0.1625	0.3028	0.3027	18.7623
4.953	21.0	42147	4.7641	0.4682	0.162	0.3015	0.3014	18.757
4.9526	22.0	44154	4.7600	0.4703	0.1626	0.3023	0.3023	18.7625
4.9571	23.0	46161	4.7592	0.4698	0.1627	0.3025	0.3025	18.781
4.9324	24.0	48168	4.7511	0.4697	0.1631	0.3022	0.3021	18.769
4.9323	25.0	50175	4.7433	0.4723	0.1649	0.304	0.3039	18.7757
4.9381	26.0	52182	4.7378	0.4703	0.1629	0.3026	0.3026	18.7782
4.9288	27.0	54189	4.7454	0.4709	0.1627	0.3026	0.3026	18.7777
4.9131	28.0	56196	4.7222	0.471	0.1652	0.3037	0.3037	18.782
4.9005	29.0	58203	4.7241	0.4719	0.1638	0.3039	0.3038	18.778
4.9051	30.0	60210	4.7225	0.4715	0.1647	0.3037	0.3036	18.7668
4.8816	31.0	62217	4.7181	0.4701	0.1631	0.3029	0.3029	18.7416
4.8687	32.0	64224	4.7061	0.4705	0.1643	0.3032	0.3031	18.7625
4.8935	33.0	66231	4.7063	0.4697	0.1632	0.3028	0.3028	18.7458
4.88	34.0	68238	4.6984	0.471	0.164	0.3039	0.3039	18.7663
4.8473	35.0	70245	4.6934	0.4699	0.1636	0.3034	0.3033	18.7531
4.8613	36.0	72252	4.6863	0.4705	0.1631	0.303	0.303	18.7797
4.8491	37.0	74259	4.6847	0.4703	0.1638	0.3037	0.3037	18.78
4.8239	38.0	76266	4.6804	0.4707	0.1632	0.3032	0.3032	18.7802
4.8767	39.0	78273	4.6788	0.4703	0.1637	0.3027	0.3026	18.7446
4.8402	40.0	80280	4.6700	0.4699	0.1633	0.3028	0.3028	18.7516
4.8261	41.0	82287	4.6660	0.4699	0.1633	0.3029	0.3028	18.7369
4.8193	42.0	84294	4.6693	0.4711	0.1654	0.3039	0.3038	18.7421
4.8161	43.0	86301	4.6636	0.4707	0.1642	0.303	0.303	18.7595
4.832	44.0	88308	4.6619	0.4708	0.1646	0.3036	0.3035	18.7423
4.8304	45.0	90315	4.6575	0.4711	0.1651	0.3038	0.3037	18.7354
4.7958	46.0	92322	4.6543	0.4711	0.165	0.3032	0.3032	18.7189
4.804	47.0	94329	4.6541	0.4711	0.1656	0.3037	0.3036	18.7396
4.7968	48.0	96336	4.6495	0.4709	0.165	0.3034	0.3034	18.7411
4.7912	49.0	98343	4.6471	0.4718	0.1655	0.3041	0.3042	18.7361
4.7721	50.0	100350	4.6469	0.4723	0.1667	0.3047	0.3047	18.7309
4.7828	51.0	102357	4.6476	0.4712	0.1656	0.3044	0.3045	18.7446
4.7934	52.0	104364	4.6453	0.4707	0.1645	0.3035	0.3035	18.7329
4.7724	53.0	106371	4.6425	0.4715	0.1657	0.304	0.304	18.7403
4.7804	54.0	108378	4.6362	0.4711	0.1658	0.3041	0.3041	18.7488
4.792	55.0	110385	4.6363	0.4706	0.1653	0.3038	0.3038	18.7281
4.7528	56.0	112392	4.6357	0.4724	0.1667	0.3044	0.3044	18.7463
4.7849	57.0	114399	4.6346	0.472	0.1661	0.3041	0.304	18.7431
4.7618	58.0	116406	4.6332	0.472	0.167	0.3046	0.3046	18.7336
4.7841	59.0	118413	4.6287	0.4716	0.1664	0.3043	0.3043	18.7369
4.7764	60.0	120420	4.6316	0.473	0.1666	0.3048	0.3047	18.7548
4.7504	61.0	122427	4.6276	0.4721	0.1671	0.3043	0.3044	18.7371
4.7629	62.0	124434	4.6250	0.4726	0.167	0.3046	0.3046	18.76
4.7764	63.0	126441	4.6264	0.4725	0.1666	0.3044	0.3044	18.7446
4.7524	64.0	128448	4.6275	0.4719	0.166	0.3041	0.3041	18.7428
4.7641	65.0	130455	4.6288	0.4728	0.1669	0.3049	0.3049	18.7458

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3