mpnet-base-airlines-news-multi-label

This model is a fine-tuned version of microsoft/mpnet-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2601
• F1: 0.8921
• Roc Auc: 0.6253

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: 7e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 65

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc
No log	1.0	57	0.3852	0.8161	0.5
No log	2.0	114	0.3612	0.8161	0.5
No log	3.0	171	0.3569	0.8161	0.5
No log	4.0	228	0.3515	0.8161	0.5
No log	5.0	285	0.3453	0.8161	0.5
No log	6.0	342	0.3403	0.8161	0.5
No log	7.0	399	0.3345	0.8161	0.5
No log	8.0	456	0.3292	0.8161	0.5
0.3585	9.0	513	0.3252	0.8161	0.5
0.3585	10.0	570	0.3175	0.8161	0.5
0.3585	11.0	627	0.3129	0.8161	0.5
0.3585	12.0	684	0.3076	0.8351	0.5029
0.3585	13.0	741	0.3024	0.8425	0.5109
0.3585	14.0	798	0.2995	0.8516	0.5163
0.3585	15.0	855	0.2953	0.8528	0.5221
0.3585	16.0	912	0.2904	0.8744	0.5426
0.3585	17.0	969	0.2875	0.8738	0.5451
0.2943	18.0	1026	0.2835	0.8833	0.5798
0.2943	19.0	1083	0.2811	0.8799	0.5710
0.2943	20.0	1140	0.2786	0.8815	0.5873
0.2943	21.0	1197	0.2761	0.8815	0.5873
0.2943	22.0	1254	0.2750	0.8838	0.5906
0.2943	23.0	1311	0.2705	0.8905	0.6194
0.2943	24.0	1368	0.2687	0.8911	0.6224
0.2943	25.0	1425	0.2674	0.8895	0.6165
0.2943	26.0	1482	0.2652	0.8911	0.6224
0.2666	27.0	1539	0.2642	0.8911	0.6224
0.2666	28.0	1596	0.2634	0.8903	0.6194
0.2666	29.0	1653	0.2612	0.8903	0.6194
0.2666	30.0	1710	0.2601	0.8921	0.6253
0.2666	31.0	1767	0.2583	0.8913	0.6328
0.2666	32.0	1824	0.2568	0.8864	0.6319
0.2666	33.0	1881	0.2563	0.8861	0.6319
0.2666	34.0	1938	0.2552	0.8869	0.6349
0.2666	35.0	1995	0.2544	0.8884	0.6378
0.2516	36.0	2052	0.2530	0.8875	0.6374
0.2516	37.0	2109	0.2523	0.8876	0.6374
0.2516	38.0	2166	0.2514	0.8889	0.6432
0.2516	39.0	2223	0.2504	0.8874	0.6453
0.2516	40.0	2280	0.2502	0.8892	0.6432
0.2516	41.0	2337	0.2495	0.8862	0.6419
0.2516	42.0	2394	0.2490	0.8867	0.6445
0.2516	43.0	2451	0.2491	0.8859	0.6365
0.2442	44.0	2508	0.2480	0.8906	0.6511
0.2442	45.0	2565	0.2476	0.8894	0.6457
0.2442	46.0	2622	0.2476	0.8888	0.6478
0.2442	47.0	2679	0.2474	0.8906	0.6511
0.2442	48.0	2736	0.2462	0.8890	0.6507
0.2442	49.0	2793	0.2461	0.8920	0.6545
0.2442	50.0	2850	0.2455	0.8894	0.6532
0.2442	51.0	2907	0.2457	0.8897	0.6507
0.2442	52.0	2964	0.2452	0.8894	0.6532
0.238	53.0	3021	0.2449	0.8903	0.6536
0.238	54.0	3078	0.2447	0.8894	0.6532
0.238	55.0	3135	0.2446	0.8894	0.6532
0.238	56.0	3192	0.2446	0.8904	0.6536
0.238	57.0	3249	0.2443	0.8894	0.6532
0.238	58.0	3306	0.2441	0.8894	0.6532
0.238	59.0	3363	0.2440	0.8911	0.6566
0.238	60.0	3420	0.2440	0.8911	0.6566
0.238	61.0	3477	0.2439	0.8903	0.6536
0.2353	62.0	3534	0.2437	0.8911	0.6566
0.2353	63.0	3591	0.2438	0.8911	0.6566
0.2353	64.0	3648	0.2437	0.8911	0.6566
0.2353	65.0	3705	0.2437	0.8911	0.6566

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1