flan-t5-rouge-squad-qg-testd

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

Loss: 0.3172
Rouge1: 0.3825
Rouge2: 0.1273
Rougel: 0.3525
Rougelsum: 0.3638

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: 0.0001
train_batch_size: 80
eval_batch_size: 80
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 320
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 320

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
62.247	1.0	3	29.8148	0.0750	0.0208	0.0666	0.0670
51.173	2.0	6	23.8876	0.0668	0.0192	0.0606	0.0610
42.1769	3.0	9	18.4541	0.0526	0.0175	0.0497	0.0495
33.256	4.0	12	12.5653	0.0824	0.0407	0.0826	0.0828
25.1048	5.0	15	8.1533	0.0710	0.0374	0.0703	0.0704
19.8539	6.0	18	7.4179	0.0661	0.0342	0.0659	0.0659
16.7613	7.0	21	6.9241	0.0793	0.0345	0.0760	0.0773
14.5402	8.0	24	5.7677	0.0895	0.0402	0.0822	0.0849
12.564	9.0	27	4.8883	0.0996	0.0445	0.0851	0.0915
11.4386	10.0	30	4.7589	0.1057	0.0450	0.0938	0.0980
10.4913	11.0	33	4.5215	0.0988	0.0410	0.0899	0.0937
9.8507	12.0	36	4.3154	0.1055	0.0444	0.0937	0.0979
9.3569	13.0	39	4.1541	0.1331	0.0567	0.1163	0.1224
8.929	14.0	42	4.0030	0.1275	0.0564	0.1116	0.1173
8.5651	15.0	45	3.8319	0.1559	0.0641	0.1310	0.1396
8.1504	16.0	48	3.6324	0.1498	0.0522	0.1246	0.1324
7.8348	17.0	51	3.4408	0.1193	0.0437	0.1000	0.1064
7.4954	18.0	54	3.2865	0.1104	0.0453	0.0949	0.1011
7.1529	19.0	57	3.1646	0.1040	0.0405	0.0888	0.0949
6.8335	20.0	60	3.0580	0.1235	0.0442	0.0995	0.1060
6.5663	21.0	63	2.9471	0.1265	0.0404	0.1021	0.1096
6.365	22.0	66	2.8180	0.1125	0.0382	0.0929	0.1000
6.1181	23.0	69	2.6687	0.1233	0.0402	0.1035	0.1098
5.8384	24.0	72	2.5123	0.1178	0.0380	0.0961	0.1006
5.5936	25.0	75	2.3656	0.1202	0.0367	0.1024	0.1070
5.3808	26.0	78	2.2328	0.1247	0.0435	0.1071	0.1118
5.1353	27.0	81	2.1150	0.1488	0.0504	0.1311	0.1352
4.9245	28.0	84	2.0090	0.1846	0.0698	0.1601	0.1734
4.7075	29.0	87	1.9051	0.2689	0.0956	0.2361	0.2553
4.5006	30.0	90	1.7932	0.2864	0.0999	0.2538	0.2738
4.3009	31.0	93	1.6756	0.2951	0.1025	0.2651	0.2847
4.0895	32.0	96	1.5604	0.3277	0.1117	0.2942	0.3130
3.8926	33.0	99	1.4551	0.3300	0.1112	0.3031	0.3185
3.7328	34.0	102	1.3608	0.3483	0.1159	0.3172	0.3347
3.51	35.0	105	1.2774	0.3560	0.1184	0.3248	0.3422
3.3655	36.0	108	1.2001	0.3635	0.1213	0.3275	0.3504
3.2056	37.0	111	1.1217	0.3592	0.1193	0.3256	0.3456
3.0361	38.0	114	1.0424	0.3560	0.1173	0.3217	0.3426
2.8543	39.0	117	0.9668	0.3572	0.1171	0.3218	0.3437
2.7049	40.0	120	0.8994	0.3594	0.1181	0.3246	0.3469
2.6019	41.0	123	0.8394	0.3594	0.1181	0.3246	0.3469
2.4523	42.0	126	0.7840	0.3628	0.1179	0.3269	0.3508
2.3234	43.0	129	0.7356	0.3651	0.1186	0.3295	0.3530
2.2271	44.0	132	0.6886	0.3675	0.1170	0.3291	0.3533
2.0893	45.0	135	0.6449	0.3657	0.1166	0.3274	0.3511
2.0187	46.0	138	0.6045	0.3657	0.1166	0.3274	0.3511
1.914	47.0	141	0.5696	0.3657	0.1166	0.3274	0.3511
1.761	48.0	144	0.5373	0.3657	0.1168	0.3274	0.3511
1.7569	49.0	147	0.5078	0.3657	0.1160	0.3269	0.3517
1.617	50.0	150	0.4828	0.3658	0.1162	0.3274	0.3515
1.5393	51.0	153	0.4599	0.3655	0.1191	0.3297	0.3517
1.4776	52.0	156	0.4384	0.3682	0.1216	0.3335	0.3550
1.4251	53.0	159	0.4202	0.3711	0.1226	0.3357	0.3576
1.3735	54.0	162	0.4040	0.3750	0.1258	0.3385	0.3602
1.3235	55.0	165	0.3889	0.3733	0.1257	0.3375	0.3580
1.265	56.0	168	0.3757	0.3733	0.1257	0.3375	0.3580
1.1689	57.0	171	0.3634	0.3733	0.1257	0.3375	0.3580
1.1205	58.0	174	0.3526	0.3737	0.1259	0.3378	0.3583
1.0933	59.0	177	0.3433	0.3743	0.1282	0.3370	0.3598
1.1121	60.0	180	0.3350	0.3782	0.1292	0.3402	0.3635
1.0716	61.0	183	0.3277	0.3601	0.1188	0.3265	0.3455
0.9798	62.0	186	0.3213	0.3550	0.1163	0.3242	0.3411
0.952	63.0	189	0.3158	0.3591	0.1226	0.3283	0.3452
0.9912	64.0	192	0.3106	0.3609	0.1222	0.3345	0.3481
0.9127	65.0	195	0.3055	0.3564	0.1204	0.3294	0.3425
0.8917	66.0	198	0.3012	0.3560	0.1201	0.3296	0.3429
0.8884	67.0	201	0.2978	0.3529	0.1189	0.3258	0.3394
0.847	68.0	204	0.2949	0.3547	0.1208	0.3278	0.3414
0.8606	69.0	207	0.2922	0.3610	0.1204	0.3321	0.3446
0.8424	70.0	210	0.2896	0.3915	0.1375	0.3564	0.3738
0.7787	71.0	213	0.2872	0.3998	0.1419	0.3630	0.3848
0.7432	72.0	216	0.2852	0.3948	0.1382	0.3590	0.3810
0.7564	73.0	219	0.2833	0.3968	0.1380	0.3613	0.3833
0.754	74.0	222	0.2816	0.3993	0.1408	0.3644	0.3851
0.7192	75.0	225	0.2802	0.4072	0.1458	0.3654	0.3923
0.7213	76.0	228	0.2790	0.4043	0.1427	0.3659	0.3885
0.6866	77.0	231	0.2779	0.4038	0.1431	0.3668	0.3874
0.6775	78.0	234	0.2769	0.4038	0.1431	0.3668	0.3874
0.6183	79.0	237	0.2759	0.4043	0.1434	0.3690	0.3881
0.6822	80.0	240	0.2750	0.3983	0.1411	0.3630	0.3821
0.6479	81.0	243	0.2743	0.4041	0.1446	0.3661	0.3870
0.6156	82.0	246	0.2737	0.4041	0.1423	0.3647	0.3870
0.6385	83.0	249	0.2732	0.4051	0.1422	0.3653	0.3877
0.5933	84.0	252	0.2727	0.4041	0.1415	0.3647	0.3882
0.5804	85.0	255	0.2724	0.3986	0.1402	0.3604	0.3830
0.5972	86.0	258	0.2724	0.3986	0.1402	0.3604	0.3830
0.5974	87.0	261	0.2726	0.3992	0.1372	0.3593	0.3821
0.5638	88.0	264	0.2728	0.4019	0.1419	0.3624	0.3830
0.5944	89.0	267	0.2728	0.3988	0.1401	0.3600	0.3811
0.5376	90.0	270	0.2727	0.3992	0.1442	0.3636	0.3807
0.5403	91.0	273	0.2725	0.4021	0.1432	0.3677	0.3821
0.5554	92.0	276	0.2727	0.3902	0.1405	0.3566	0.3718
0.5088	93.0	279	0.2732	0.3857	0.1374	0.3545	0.3686
0.5104	94.0	282	0.2736	0.3871	0.1429	0.3553	0.3687
0.5169	95.0	285	0.2738	0.3896	0.1446	0.3573	0.3716
0.5073	96.0	288	0.2744	0.3940	0.1407	0.3585	0.3770
0.493	97.0	291	0.2750	0.3936	0.1401	0.3583	0.3765
0.5112	98.0	294	0.2756	0.3934	0.1400	0.3582	0.3763
0.4956	99.0	297	0.2760	0.3924	0.1400	0.3577	0.3755
0.451	100.0	300	0.2762	0.3946	0.1408	0.3605	0.3785
0.4518	101.0	303	0.2766	0.3931	0.1402	0.3584	0.3762
0.4978	102.0	306	0.2772	0.3930	0.1411	0.3589	0.3764
0.4707	103.0	309	0.2782	0.3899	0.1411	0.3589	0.3727
0.462	104.0	312	0.2793	0.3899	0.1411	0.3589	0.3727
0.4706	105.0	315	0.2799	0.3955	0.1367	0.3616	0.3777
0.4762	106.0	318	0.2807	0.3918	0.1343	0.3588	0.3741
0.4111	107.0	321	0.2811	0.3918	0.1343	0.3588	0.3741
0.417	108.0	324	0.2815	0.4029	0.1406	0.3685	0.3846
0.4255	109.0	327	0.2819	0.3983	0.1350	0.3617	0.3817
0.4114	110.0	330	0.2829	0.3977	0.1383	0.3594	0.3800
0.4327	111.0	333	0.2840	0.3971	0.1376	0.3588	0.3794
0.4261	112.0	336	0.2847	0.3892	0.1322	0.3505	0.3717
0.4185	113.0	339	0.2852	0.3792	0.1228	0.3406	0.3633
0.4145	114.0	342	0.2859	0.3794	0.1228	0.3411	0.3639
0.4198	115.0	345	0.2867	0.3757	0.1184	0.3373	0.3602
0.4012	116.0	348	0.2874	0.3806	0.1203	0.3407	0.3638
0.4371	117.0	351	0.2878	0.3780	0.1209	0.3376	0.3605
0.4001	118.0	354	0.2880	0.3779	0.1207	0.3375	0.3604
0.3914	119.0	357	0.2885	0.3835	0.1277	0.3435	0.3655
0.3985	120.0	360	0.2891	0.3739	0.1221	0.3381	0.3559
0.3902	121.0	363	0.2902	0.3825	0.1226	0.3431	0.3649
0.4109	122.0	366	0.2914	0.3825	0.1226	0.3431	0.3649
0.3785	123.0	369	0.2921	0.3790	0.1223	0.3393	0.3611
0.3985	124.0	372	0.2926	0.3790	0.1223	0.3393	0.3611
0.3709	125.0	375	0.2928	0.3768	0.1215	0.3376	0.3593
0.3844	126.0	378	0.2933	0.3769	0.1196	0.3366	0.3595
0.3716	127.0	381	0.2938	0.3769	0.1196	0.3366	0.3595
0.3907	128.0	384	0.2945	0.3789	0.1205	0.3380	0.3609
0.3565	129.0	387	0.2951	0.3811	0.1219	0.3388	0.3629
0.363	130.0	390	0.2959	0.3794	0.1211	0.3374	0.3616
0.3389	131.0	393	0.2967	0.3798	0.1234	0.3377	0.3611
0.3862	132.0	396	0.2975	0.3804	0.1234	0.3417	0.3609
0.3791	133.0	399	0.2982	0.3804	0.1234	0.3417	0.3609
0.3707	134.0	402	0.2986	0.3770	0.1210	0.3388	0.3573
0.3381	135.0	405	0.2989	0.3742	0.1192	0.3369	0.3556
0.3637	136.0	408	0.2994	0.3815	0.1270	0.3457	0.3616
0.3488	137.0	411	0.2999	0.3819	0.1268	0.3461	0.3620
0.3447	138.0	414	0.3003	0.3819	0.1268	0.3461	0.3620
0.3503	139.0	417	0.3005	0.3821	0.1270	0.3463	0.3622
0.3337	140.0	420	0.3008	0.3775	0.1244	0.3429	0.3577
0.3543	141.0	423	0.3013	0.3792	0.1254	0.3441	0.3590
0.3206	142.0	426	0.3017	0.3796	0.1254	0.3445	0.3593
0.3527	143.0	429	0.3021	0.3781	0.1257	0.3450	0.3590
0.3393	144.0	432	0.3026	0.3825	0.1313	0.3507	0.3635
0.3653	145.0	435	0.3032	0.3821	0.1304	0.3511	0.3634
0.3169	146.0	438	0.3039	0.3822	0.1296	0.3517	0.3637
0.3539	147.0	441	0.3042	0.3806	0.1295	0.3493	0.3618
0.3131	148.0	444	0.3047	0.3806	0.1295	0.3493	0.3618
0.3501	149.0	447	0.3053	0.3811	0.1295	0.3497	0.3622
0.3273	150.0	450	0.3058	0.3766	0.1248	0.3438	0.3578
0.3397	151.0	453	0.3061	0.3766	0.1248	0.3438	0.3578
0.3215	152.0	456	0.3062	0.3764	0.1239	0.3440	0.3577
0.3169	153.0	459	0.3064	0.3764	0.1239	0.3440	0.3577
0.3411	154.0	462	0.3067	0.3764	0.1239	0.3440	0.3577
0.3145	155.0	465	0.3069	0.3777	0.1251	0.3453	0.3588
0.3356	156.0	468	0.3072	0.3755	0.1240	0.3430	0.3565
0.3088	157.0	471	0.3073	0.3760	0.1241	0.3434	0.3569
0.3266	158.0	474	0.3077	0.3760	0.1241	0.3434	0.3569
0.3275	159.0	477	0.3082	0.3776	0.1237	0.3430	0.3573
0.3328	160.0	480	0.3086	0.3776	0.1237	0.3430	0.3573
0.3192	161.0	483	0.3090	0.3790	0.1246	0.3435	0.3576
0.3205	162.0	486	0.3094	0.3795	0.1246	0.3439	0.3581
0.3099	163.0	489	0.3098	0.3800	0.1248	0.3443	0.3586
0.3239	164.0	492	0.3101	0.3800	0.1248	0.3443	0.3586
0.2915	165.0	495	0.3104	0.3777	0.1260	0.3447	0.3584
0.3374	166.0	498	0.3107	0.3777	0.1260	0.3447	0.3584
0.3167	167.0	501	0.3113	0.3781	0.1242	0.3473	0.3596
0.3076	168.0	504	0.3116	0.3781	0.1242	0.3473	0.3596
0.3128	169.0	507	0.3118	0.3762	0.1240	0.3463	0.3579
0.3082	170.0	510	0.3120	0.3762	0.1228	0.3458	0.3583
0.3139	171.0	513	0.3120	0.3796	0.1231	0.3479	0.3600
0.3136	172.0	516	0.3119	0.3796	0.1231	0.3479	0.3600
0.3149	173.0	519	0.3119	0.3754	0.1219	0.3432	0.3562
0.3149	174.0	522	0.3121	0.3736	0.1203	0.3417	0.3550
0.3192	175.0	525	0.3121	0.3773	0.1215	0.3439	0.3578
0.3136	176.0	528	0.3123	0.3773	0.1215	0.3439	0.3578
0.3068	177.0	531	0.3125	0.3773	0.1215	0.3439	0.3578
0.3083	178.0	534	0.3129	0.3755	0.1206	0.3426	0.3567
0.3096	179.0	537	0.3133	0.3755	0.1206	0.3426	0.3567
0.3081	180.0	540	0.3137	0.3767	0.1253	0.3456	0.3571
0.3131	181.0	543	0.3140	0.3767	0.1253	0.3456	0.3571
0.2881	182.0	546	0.3142	0.3773	0.1255	0.3460	0.3576
0.2857	183.0	549	0.3143	0.3773	0.1255	0.3460	0.3576
0.2978	184.0	552	0.3145	0.3773	0.1255	0.3460	0.3576
0.2917	185.0	555	0.3146	0.3773	0.1255	0.3460	0.3576
0.3251	186.0	558	0.3147	0.3773	0.1255	0.3460	0.3576
0.3096	187.0	561	0.3147	0.3773	0.1255	0.3460	0.3576
0.2918	188.0	564	0.3149	0.3773	0.1255	0.3460	0.3576
0.2953	189.0	567	0.3152	0.3773	0.1255	0.3460	0.3576
0.2976	190.0	570	0.3154	0.3782	0.1217	0.3467	0.3593
0.3221	191.0	573	0.3154	0.3782	0.1217	0.3467	0.3593
0.285	192.0	576	0.3155	0.3760	0.1208	0.3430	0.3571
0.3	193.0	579	0.3156	0.3760	0.1208	0.3430	0.3571
0.2962	194.0	582	0.3159	0.3760	0.1208	0.3430	0.3571
0.3068	195.0	585	0.3161	0.3760	0.1208	0.3430	0.3571
0.3138	196.0	588	0.3162	0.3760	0.1208	0.3430	0.3571
0.305	197.0	591	0.3163	0.3760	0.1208	0.3430	0.3571
0.3013	198.0	594	0.3164	0.3760	0.1208	0.3430	0.3571
0.2877	199.0	597	0.3166	0.3760	0.1208	0.3430	0.3571
0.2951	200.0	600	0.3167	0.3760	0.1208	0.3430	0.3571
0.316	201.0	603	0.3169	0.3762	0.1210	0.3432	0.3573
0.2974	202.0	606	0.3170	0.3783	0.1218	0.3469	0.3595
0.2773	203.0	609	0.3171	0.3783	0.1218	0.3469	0.3595
0.2948	204.0	612	0.3171	0.3783	0.1218	0.3469	0.3595
0.3053	205.0	615	0.3171	0.3783	0.1218	0.3469	0.3595
0.2965	206.0	618	0.3172	0.3783	0.1218	0.3469	0.3595
0.2936	207.0	621	0.3172	0.3825	0.1273	0.3525	0.3638
0.2952	208.0	624	0.3172	0.3825	0.1273	0.3525	0.3638
0.293	209.0	627	0.3172	0.3825	0.1273	0.3525	0.3638
0.2994	210.0	630	0.3172	0.3825	0.1273	0.3525	0.3638
0.3017	211.0	633	0.3172	0.3825	0.1273	0.3525	0.3638
0.3058	212.0	636	0.3172	0.3825	0.1273	0.3525	0.3638
0.2976	213.0	639	0.3172	0.3825	0.1273	0.3525	0.3638
0.5935	213.4	640	0.3172	0.3825	0.1273	0.3525	0.3638

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0

devagonal/flan-t5-rouge-squad-qg-testd

flan-t5-rouge-squad-qg-testd

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种