TestZee/t5-small-finetuned-pytorch-final

来源：🤗🏴‍☠️ 救援者：Alogotronapache-2.0245 MBother✓ 4 个校验和1,465 天前更新

t5-small-finetuned-pytorch-final

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

Loss: 1.6589
Rouge1: 25.8199
Rouge2: 14.6736
Rougel: 22.3682
Rougelsum: 23.917
Gen Len: 19.0

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: 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: 35
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.0989	1.0	503	1.8152	24.4365	13.3113	20.9862	22.6351	19.0
1.9704	2.0	1006	1.7720	24.8843	13.7084	21.4551	22.9787	19.0
1.9241	3.0	1509	1.7468	25.1988	14.052	21.7041	23.2981	19.0
1.8856	4.0	2012	1.7268	25.502	14.3573	21.9449	23.5741	19.0
1.8653	5.0	2515	1.7156	25.6461	14.3872	22.0824	23.6502	19.0
1.8367	6.0	3018	1.7057	25.6513	14.4842	22.2314	23.7009	19.0
1.8183	7.0	3521	1.6993	25.6377	14.481	22.185	23.6855	19.0
1.8068	8.0	4024	1.6945	25.5275	14.3184	22.03	23.5524	19.0
1.7959	9.0	4527	1.6885	25.4232	14.2443	21.9691	23.4711	19.0
1.7741	10.0	5030	1.6840	25.5169	14.2654	22.0518	23.5864	19.0
1.7665	11.0	5533	1.6817	25.5237	14.3758	22.094	23.5891	19.0
1.7541	12.0	6036	1.6779	25.2572	14.1939	21.816	23.3577	19.0
1.7479	13.0	6539	1.6761	25.3922	14.4173	22.0299	23.5163	19.0
1.7308	14.0	7042	1.6742	25.3631	14.2906	22.0221	23.5128	19.0
1.7261	15.0	7545	1.6717	25.4318	14.3493	22.0454	23.5278	19.0
1.7181	16.0	8048	1.6691	25.4043	14.325	22.0252	23.5423	19.0
1.7048	17.0	8551	1.6691	25.6406	14.5424	22.2377	23.7325	19.0
1.7064	18.0	9054	1.6671	25.4986	14.3177	22.0629	23.5943	19.0
1.7003	19.0	9557	1.6687	25.6196	14.4546	22.2079	23.7184	19.0
1.6858	20.0	10060	1.6660	25.6864	14.5874	22.3071	23.8151	19.0
1.6861	21.0	10563	1.6648	25.6698	14.5281	22.2717	23.797	19.0
1.684	22.0	11066	1.6635	25.7104	14.5393	22.2573	23.829	19.0
1.6751	23.0	11569	1.6615	25.7254	14.5923	22.2509	23.8439	19.0
1.6741	24.0	12072	1.6624	25.7821	14.663	22.3164	23.8809	19.0
1.6765	25.0	12575	1.6621	25.7689	14.5796	22.2779	23.8765	19.0
1.6562	26.0	13078	1.6616	25.7856	14.6224	22.3298	23.9215	19.0
1.6636	27.0	13581	1.6610	25.83	14.6569	22.4229	23.9404	19.0
1.671	28.0	14084	1.6599	25.6857	14.5126	22.2093	23.788	19.0
1.6467	29.0	14587	1.6602	25.8111	14.6784	22.3599	23.9132	19.0
1.6556	30.0	15090	1.6594	25.8887	14.7244	22.3998	23.9739	19.0
1.6463	31.0	15593	1.6594	25.8534	14.6966	22.3867	23.9439	19.0
1.6548	32.0	16096	1.6600	25.839	14.674	22.3763	23.9438	19.0
1.6458	33.0	16599	1.6590	25.8364	14.6589	22.3512	23.9243	19.0
1.6431	34.0	17102	1.6590	25.8314	14.6733	22.3526	23.9288	19.0
1.6637	35.0	17605	1.6589	25.8199	14.6736	22.3682	23.917	19.0

Framework versions

Transformers 4.17.0
Pytorch 1.12.1+cu113
Datasets 2.6.1
Tokenizers 0.13.2