finetune-led-35000

silmi224/finetune-led-35000

finetune-led-35000 is a summarization model from silmi224. Use it when you need a shorter version of a longer text. It is set up for transformers.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

2% of all-time downloads

All-time downloads

400

Public

Parameters

162M

648 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors648 MB · 99%

At a glance

Task: Summarization
Library: transformers
Model type: led
Access: Public
Created: Jul 2, 2024
Updated: Jul 2, 2024
SHA: f7c259f2

Try a prompt

Model card

From the Hugging Face model README

finetune-led-35000

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.8474
Rouge1 Precision: 0.2576
Rouge1 Recall: 0.3438
Rouge1 Fmeasure: 0.2911

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
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1 Fmeasure	Rouge1 Precision	Rouge1 Recall
2.0001	0.01	10	2.8444	0.2732	0.2484	0.3213
1.7409	0.01	20	2.6895	0.2682	0.2375	0.3203
1.6096	0.02	30	2.6335	0.2797	0.246	0.3362
1.7452	0.03	40	2.5817	0.285	0.251	0.3417
1.6212	0.03	50	2.5765	0.2899	0.253	0.3515
1.6743	0.04	60	2.4425	0.2705	0.241	0.3188
1.5229	0.04	70	2.5146	0.2802	0.2501	0.3294
1.5681	0.05	80	2.4785	0.2665	0.2302	0.3273
1.5811	0.06	90	2.3622	0.287	0.2577	0.3349
1.5642	0.06	100	2.4053	0.2779	0.2449	0.3322
1.5175	0.07	110	2.3818	0.2862	0.2577	0.3324
1.5073	0.08	120	2.3930	0.2905	0.25	0.3587
1.5336	0.12	130	2.3619	0.2755	0.2456	0.3242
1.5155	0.13	140	2.3696	0.2752	0.243	0.3284
1.5124	0.14	150	2.3240	0.2788	0.2478	0.3296
1.5187	0.15	160	2.2883	0.2739	0.2438	0.3225
1.4642	0.16	170	2.3006	0.2769	0.2442	0.3307
1.6535	0.16	180	2.2766	0.2716	0.2435	0.3165
1.4924	0.17	190	2.3077	0.2782	0.2463	0.3303
1.5221	0.18	200	2.2725	0.2829	0.2502	0.3357
1.388	0.19	210	2.2555	0.2811	0.2502	0.3305
1.5172	0.2	220	2.2519	0.2758	0.2458	0.3243
1.498	0.21	230	2.2585	0.2809	0.2428	0.3437
1.447	0.22	240	2.2110	0.2745	0.2457	0.3206
1.4637	0.23	250	2.2511	0.2854	0.252	0.3395
1.5084	0.24	260	2.2070	0.2876	0.2547	0.341
1.4347	0.25	270	2.2558	0.2852	0.2493	0.344
1.3108	0.26	280	2.2498	0.2815	0.2488	0.3349
1.3894	0.27	290	2.2302	0.2794	0.2454	0.3347
1.4392	0.27	300	2.2021	0.276	0.2412	0.3327
1.3774	0.28	310	2.1958	0.2763	0.244	0.3286
1.3696	0.29	320	2.1657	0.2811	0.2512	0.3292
1.4186	0.3	330	2.1533	0.2807	0.251	0.3288
1.3073	0.31	340	2.1890	0.2805	0.2486	0.3324
1.3533	0.32	350	2.1639	0.2836	0.2546	0.3302
1.3691	0.33	360	2.1255	0.2765	0.25	0.3188
1.3716	0.34	370	2.1436	0.2818	0.2512	0.3315
1.3985	0.35	380	2.1160	0.2859	0.2555	0.3351
1.2993	0.36	390	2.1417	0.2756	0.2442	0.3262
1.3453	0.37	400	2.1459	0.2875	0.256	0.3386
1.4036	0.37	410	2.1200	0.2811	0.2506	0.33
1.3784	0.38	420	2.0878	0.2885	0.2592	0.335
1.2923	0.39	430	2.1342	0.2873	0.2535	0.3423
1.3629	0.4	440	2.0887	0.2848	0.2546	0.3328
1.4151	0.41	450	2.0905	0.2857	0.254	0.3365
1.3679	0.42	460	2.0829	0.285	0.2525	0.3375
1.339	0.43	470	2.0786	0.2791	0.2452	0.334
1.4258	0.44	480	2.0726	0.2877	0.2603	0.3317
1.4056	0.45	490	2.0995	0.2891	0.2556	0.3428
1.3548	0.46	500	2.0637	0.2757	0.2481	0.3199
1.3253	0.47	510	2.0638	0.2794	0.25	0.3266
1.264	0.48	520	2.0609	0.2861	0.2587	0.3296
1.3307	0.48	530	2.0396	0.2823	0.2559	0.3243
1.3536	0.49	540	2.0464	0.2824	0.2532	0.3288
1.2592	0.5	550	2.0480	0.2876	0.2592	0.3333
1.358	0.51	560	2.0432	0.2818	0.2528	0.3284
1.3227	0.52	570	2.0560	0.2831	0.2502	0.3365
1.3189	0.53	580	2.0311	0.2823	0.251	0.3321
1.3367	0.54	590	2.0498	0.285	0.2538	0.335
1.3473	0.55	600	2.0690	0.2773	0.2452	0.3292
1.2846	0.56	610	2.0555	0.2796	0.2473	0.3321
1.3066	0.57	620	2.0684	0.2799	0.245	0.3366
1.3193	0.58	630	2.0467	0.2852	0.2536	0.336
1.269	0.59	640	2.0381	0.2859	0.2561	0.3337
1.2906	0.59	650	2.0191	0.2831	0.2514	0.3338
1.2981	0.6	660	2.0184	0.2783	0.249	0.3251
1.2888	0.61	670	2.0295	0.2827	0.2515	0.3331
1.3179	0.62	680	2.0121	0.2885	0.2611	0.333
1.3313	0.63	690	2.0296	0.2739	0.2427	0.3245
1.1749	0.64	700	2.0419	0.2809	0.2507	0.3298
1.3023	0.65	710	2.0275	0.2838	0.2504	0.3379
1.262	0.66	720	1.9974	0.286	0.2539	0.3378
1.2906	0.67	730	1.9839	0.2839	0.252	0.3357
1.24	0.68	740	2.0041	0.286	0.2528	0.3401
1.239	0.69	750	2.0116	0.2789	0.2455	0.3326
1.1972	0.69	760	2.0293	0.2861	0.2536	0.3385
1.2114	0.7	770	2.0271	0.2738	0.2436	0.322
1.2711	0.71	780	2.0084	0.2881	0.2548	0.3417
1.262	0.72	790	1.9984	0.2806	0.2488	0.3322
1.2616	0.73	800	1.9715	0.2856	0.2541	0.3364
1.2765	0.74	810	1.9718	0.2825	0.2494	0.3356
1.2151	0.75	820	1.9947	0.2857	0.2513	0.341
1.3165	0.76	830	1.9854	0.2863	0.2524	0.3411
1.2704	0.77	840	1.9858	0.2903	0.2569	0.3443
1.3032	0.78	850	1.9774	0.2926	0.2583	0.3481
1.2461	0.79	860	1.9596	0.2847	0.2556	0.3314
1.2288	0.8	870	1.9873	0.2868	0.2547	0.339
1.2278	0.8	880	1.9712	0.289	0.2546	0.3455
1.2119	0.81	890	1.9862	0.2822	0.2478	0.338
1.3363	0.82	900	1.9555	0.2871	0.2576	0.3349
1.2324	0.83	910	1.9394	0.2878	0.2588	0.3339
1.2528	0.84	920	1.9593	0.2801	0.2498	0.3289
1.2572	0.85	930	1.9500	0.2825	0.2507	0.3337
1.2045	0.86	940	1.9586	0.2901	0.2589	0.3401
1.2173	0.87	950	1.9551	0.281	0.2487	0.3328
1.2315	0.88	960	1.9307	0.2842	0.2533	0.3337
1.2445	0.89	970	1.9362	0.2853	0.2537	0.336
1.2491	0.9	980	1.9614	0.2829	0.2482	0.3397
1.3081	0.91	990	1.9500	0.2857	0.2513	0.3411
1.1928	0.91	1000	1.9439	0.2826	0.2514	0.333
1.2243	0.92	1010	1.9074	0.2883	0.259	0.3346
1.2662	0.93	1020	1.9143	0.2912	0.2593	0.3422
1.2223	0.94	1030	1.9342	0.2899	0.2581	0.3408
1.2499	0.95	1040	1.9352	0.2835	0.2507	0.3366
1.3395	0.96	1050	1.9284	0.2864	0.2548	0.3375
1.1908	0.97	1060	1.9471	0.2853	0.2528	0.3376
1.2473	0.98	1070	1.9462	0.2941	0.2613	0.3472
1.2139	0.99	1080	1.9317	0.2859	0.2534	0.338
1.2534	1.0	1090	1.9278	0.2938	0.2594	0.3488
1.2204	1.01	1100	1.9177	0.2912	0.2596	0.341
1.2399	1.01	1110	1.9236	0.2903	0.2568	0.3443
1.1541	1.02	1120	1.9441	0.2889	0.2548	0.3431
1.1038	1.03	1130	1.9223	0.2925	0.2626	0.3399
1.1177	1.04	1140	1.9244	0.2881	0.2565	0.338
1.1224	1.05	1150	1.9324	0.2884	0.2547	0.3428
1.104	1.06	1160	1.9188	0.2798	0.2482	0.3304
1.175	1.07	1170	1.9042	0.2915	0.2618	0.3388
1.102	1.08	1180	1.9325	0.2853	0.253	0.3372
1.0829	1.09	1190	1.9503	0.2819	0.2478	0.3371
1.1842	1.1	1200	1.9360	0.2784	0.2438	0.3346
1.1552	1.11	1210	1.9055	0.286	0.254	0.3369
1.1266	1.12	1220	1.9106	0.286	0.2555	0.3345
1.1288	1.13	1230	1.9072	0.2865	0.2566	0.3336
1.1722	1.13	1240	1.9114	0.2856	0.2539	0.3364
1.1514	1.14	1250	1.9180	0.2906	0.2561	0.3461
1.1642	1.15	1260	1.9226	0.2918	0.2571	0.3475
1.1464	1.16	1270	1.9004	0.2819	0.2525	0.3283
1.1829	1.17	1280	1.9181	0.2935	0.2568	0.3524
1.17	1.18	1290	1.9031	0.2848	0.2523	0.3369
1.0751	1.19	1300	1.9334	0.2875	0.2531	0.3428
1.1327	1.2	1310	1.8966	0.2891	0.2568	0.3407
1.1319	1.21	1320	1.9076	0.2902	0.2575	0.3422
1.106	1.22	1330	1.8941	0.2908	0.259	0.3413
1.1721	1.23	1340	1.8956	0.2945	0.2609	0.3479
1.1964	1.23	1350	1.9140	0.2851	0.2513	0.3389
1.1195	1.24	1360	1.9168	0.2917	0.2561	0.3483
1.1352	1.25	1370	1.8962	0.286	0.253	0.3389
1.1164	1.26	1380	1.9050	0.2916	0.258	0.3453
1.1219	1.27	1390	1.9054	0.2872	0.2551	0.3386
1.1571	1.28	1400	1.8845	0.2896	0.2574	0.3402
1.2033	1.29	1410	1.8985	0.2852	0.2532	0.3362
1.1114	1.3	1420	1.8956	0.2882	0.2559	0.3395
1.1268	1.31	1430	1.8955	0.2895	0.2563	0.3424
1.1347	1.32	1440	1.8883	0.2865	0.2524	0.3412
1.0345	1.33	1450	1.8960	0.2895	0.2571	0.3412
1.1231	1.34	1460	1.8873	0.29	0.2575	0.3415
1.236	1.34	1470	1.8744	0.2898	0.2578	0.34
1.1054	1.35	1480	1.8867	0.2884	0.2546	0.3425
1.1393	1.36	1490	1.8907	0.2927	0.2605	0.344
1.1004	1.37	1500	1.8953	0.288	0.2543	0.3416
1.1482	1.38	1510	1.8731	0.288	0.2568	0.3377
1.1701	1.39	1520	1.8868	0.2866	0.2525	0.3411
1.1233	1.4	1530	1.8803	0.2882	0.2562	0.3385
1.0685	1.41	1540	1.8843	0.2935	0.262	0.3433
1.0657	1.42	1550	1.8748	0.2892	0.2553	0.3437
1.1275	1.43	1560	1.8804	0.2881	0.2553	0.3405
1.0883	1.44	1570	1.8803	0.2868	0.2527	0.3412
1.1096	1.45	1580	1.8862	0.2927	0.2586	0.3472
1.1521	1.45	1590	1.8724	0.288	0.2564	0.3379
1.142	1.46	1600	1.8788	0.2926	0.2593	0.3454
1.0451	1.47	1610	1.8684	0.2863	0.2571	0.3324
1.1294	1.48	1620	1.8704	0.2902	0.2569	0.3427
1.1671	1.49	1630	1.8756	0.2909	0.259	0.3413
1.2252	1.5	1640	1.8618	0.2937	0.2599	0.347
1.0834	1.51	1650	1.8776	0.2909	0.2589	0.3416
1.0417	1.52	1660	1.8658	0.2911	0.2592	0.342
1.1036	1.53	1670	1.8789	0.289	0.2553	0.343
1.1575	1.54	1680	1.8608	0.2927	0.2597	0.3452
1.058	1.55	1690	1.8804	0.2921	0.2585	0.3455
1.1251	1.55	1700	1.8682	0.2973	0.2637	0.3503
1.0818	1.56	1710	1.8800	0.2887	0.2544	0.3432
1.1346	1.57	1720	1.8577	0.289	0.2564	0.3404
1.1024	1.58	1730	1.8681	0.2946	0.2608	0.3482
1.0823	1.59	1740	1.8603	0.2908	0.2584	0.342
1.0562	1.6	1750	1.8670	0.2931	0.2584	0.3484
1.1128	1.61	1760	1.8576	0.2926	0.2603	0.3439
1.0769	1.62	1770	1.8709	0.2902	0.2568	0.3434
1.0422	1.63	1780	1.8597	0.2911	0.2587	0.3425
1.1912	1.64	1790	1.8648	0.2911	0.2574	0.3448
1.1349	1.65	1800	1.8667	0.2933	0.2606	0.3453
1.1195	1.66	1810	1.8684	0.2899	0.2568	0.3422
1.1186	1.66	1820	1.8581	0.2908	0.2579	0.3434
1.0795	1.67	1830	1.8639	0.2907	0.2561	0.3462
1.1394	1.68	1840	1.8467	0.2929	0.2602	0.3446
1.0743	1.69	1850	1.8682	0.291	0.2585	0.3428
1.0954	1.7	1860	1.8504	0.2928	0.2603	0.3445
1.0938	1.71	1870	1.8604	0.2916	0.2589	0.3436
1.1093	1.72	1880	1.8427	0.2897	0.2581	0.3398
1.1399	1.73	1890	1.8715	0.2891	0.2561	0.3422
1.1574	1.74	1900	1.8448	0.2893	0.2568	0.3409
1.1244	1.75	1910	1.8594	0.2927	0.2597	0.3453
1.1205	1.76	1920	1.8492	0.2922	0.2606	0.3425
1.1218	1.77	1930	1.8547	0.2906	0.2591	0.3401
1.1208	1.77	1940	1.8605	0.2924	0.2588	0.3459
1.0983	1.78	1950	1.8425	0.2933	0.2611	0.3442
1.1992	1.79	1960	1.8587	0.2907	0.2565	0.3455
1.1724	1.8	1970	1.8413	0.2909	0.2576	0.3435
1.1344	1.81	1980	1.8494	0.2904	0.2583	0.3413
1.1469	1.82	1990	1.8463	0.2911	0.2581	0.3437
1.1491	1.83	2000	1.8530	0.2905	0.2568	0.3441
1.0913	1.84	2010	1.8493	0.2913	0.258	0.3443
1.1298	1.85	2020	1.8465	0.2905	0.2573	0.3433
1.1202	1.86	2030	1.8488	0.2892	0.256	0.3419
1.1439	1.87	2040	1.8494	0.2911	0.2584	0.3428
1.0328	1.87	2050	1.8469	0.2907	0.2582	0.3423
1.1347	1.88	2060	1.8426	0.29	0.2576	0.341
1.094	1.89	2070	1.8480	0.2905	0.2577	0.3425
1.1201	1.9	2080	1.8542	0.2896	0.2568	0.3415
1.1475	1.91	2090	1.8520	0.29	0.2574	0.3416
1.0793	1.92	2100	1.8506	0.2897	0.2569	0.3414
1.0669	1.93	2110	1.8484	0.2907	0.2577	0.3426
1.1276	1.94	2120	1.8487	0.2904	0.2573	0.3427
1.0902	1.95	2130	1.8487	0.2904	0.2575	0.3423
1.1449	1.96	2140	1.8490	0.2898	0.2569	0.3419
1.1142	1.97	2150	1.8505	0.29	0.2569	0.3424
1.1475	1.98	2160	1.8501	0.2895	0.2561	0.3424
1.0663	1.98	2170	1.8485	0.2906	0.2571	0.3434
1.1454	1.99	2180	1.8475	0.2907	0.2573	0.3435

Framework versions

Transformers 4.35.2
Pytorch 2.1.1+cu121
Datasets 2.14.5
Tokenizers 0.15.1

Task: Summarization
Library: transformers
Type: led
Created: Jul 2, 2024
Updated: Jul 2, 2024