Modelpublic

genki10/ASAP_FineTuningBERT_UnAugV6_k1_task1_organization_k1_k1_fold3

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes3downloads

Model Card

ASAPFineTuningBERTUnAugV6k1task1organizationk1k1fold3

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

—Loss: 0.8190
—Qwk: 0.5679
—Mse: 0.8187
—Rmse: 0.9048

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
—trainbatchsize: 64
—evalbatchsize: 64
—seed: 42
—optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	1.0	1	13.7258	0.0	13.7236	3.7045
No log	2.0	2	11.7633	-0.0003	11.7616	3.4295
No log	3.0	3	10.0506	0.0072	10.0488	3.1700
No log	4.0	4	9.0607	0.0018	9.0591	3.0098
No log	5.0	5	8.5243	0.0	8.5231	2.9194
No log	6.0	6	7.3490	0.0	7.3479	2.7107
No log	7.0	7	6.4524	0.0211	6.4508	2.5398
No log	8.0	8	5.6965	0.0443	5.6950	2.3864
No log	9.0	9	4.9700	0.0275	4.9690	2.2291
No log	10.0	10	4.5669	0.0204	4.5662	2.1369
No log	11.0	11	3.9133	0.0151	3.9126	1.9780
No log	12.0	12	3.3405	0.0076	3.3396	1.8275
No log	13.0	13	2.9568	0.0076	2.9558	1.7192
No log	14.0	14	2.5901	0.0546	2.5893	1.6091
No log	15.0	15	2.2342	0.1028	2.2334	1.4944
No log	16.0	16	2.0119	0.0720	2.0110	1.4181
No log	17.0	17	1.7407	0.0599	1.7399	1.3190
No log	18.0	18	1.5275	0.0799	1.5268	1.2356
No log	19.0	19	1.3508	0.0694	1.3502	1.1620
No log	20.0	20	1.1820	0.0488	1.1814	1.0869
No log	21.0	21	1.1396	0.0463	1.1389	1.0672
No log	22.0	22	1.0163	0.0463	1.0157	1.0078
No log	23.0	23	0.9005	0.1605	0.9002	0.9488
No log	24.0	24	1.0645	0.1165	1.0642	1.0316
No log	25.0	25	1.0320	0.1391	1.0318	1.0158
No log	26.0	26	0.8108	0.3087	0.8107	0.9004
No log	27.0	27	0.7247	0.3797	0.7245	0.8512
No log	28.0	28	0.7362	0.3678	0.7359	0.8579
No log	29.0	29	0.6807	0.3864	0.6805	0.8249
No log	30.0	30	0.6322	0.4174	0.6321	0.7951
No log	31.0	31	0.6468	0.4195	0.6468	0.8042
No log	32.0	32	0.6159	0.4396	0.6158	0.7848
No log	33.0	33	0.5927	0.4221	0.5927	0.7699
No log	34.0	34	0.6049	0.4457	0.6049	0.7777
No log	35.0	35	0.5787	0.4728	0.5787	0.7607
No log	36.0	36	0.5853	0.5196	0.5853	0.7650
No log	37.0	37	0.6013	0.5212	0.6013	0.7754
No log	38.0	38	0.5742	0.5267	0.5743	0.7578
No log	39.0	39	0.5687	0.5444	0.5688	0.7542
No log	40.0	40	0.5647	0.5650	0.5647	0.7515
No log	41.0	41	0.5653	0.5924	0.5653	0.7519
No log	42.0	42	0.5749	0.5771	0.5749	0.7582
No log	43.0	43	0.5623	0.5883	0.5624	0.7499
No log	44.0	44	0.5894	0.6144	0.5895	0.7678
No log	45.0	45	0.5904	0.6115	0.5905	0.7684
No log	46.0	46	0.5529	0.6027	0.5529	0.7436
No log	47.0	47	0.5975	0.5756	0.5976	0.7730
No log	48.0	48	0.6362	0.5696	0.6363	0.7977
No log	49.0	49	0.5909	0.5848	0.5910	0.7688
No log	50.0	50	0.5425	0.6160	0.5425	0.7366
No log	51.0	51	0.6432	0.6236	0.6432	0.8020
No log	52.0	52	0.7176	0.5877	0.7177	0.8472
No log	53.0	53	0.7014	0.6034	0.7014	0.8375
No log	54.0	54	0.6239	0.6396	0.6238	0.7898
No log	55.0	55	0.5649	0.6122	0.5649	0.7516
No log	56.0	56	0.5527	0.6167	0.5526	0.7434
No log	57.0	57	0.5565	0.6168	0.5565	0.7460
No log	58.0	58	0.6175	0.6223	0.6174	0.7858
No log	59.0	59	0.7474	0.5970	0.7473	0.8645
No log	60.0	60	0.7865	0.5876	0.7864	0.8868
No log	61.0	61	0.7278	0.6182	0.7276	0.8530
No log	62.0	62	0.6133	0.6474	0.6132	0.7831
No log	63.0	63	0.5547	0.6309	0.5546	0.7447
No log	64.0	64	0.5680	0.6137	0.5679	0.7536
No log	65.0	65	0.5668	0.6174	0.5667	0.7528
No log	66.0	66	0.5707	0.6381	0.5706	0.7554
No log	67.0	67	0.6431	0.6456	0.6430	0.8018
No log	68.0	68	0.7147	0.6353	0.7145	0.8453
No log	69.0	69	0.7924	0.6022	0.7922	0.8900
No log	70.0	70	0.7956	0.5987	0.7954	0.8919
No log	71.0	71	0.7354	0.6129	0.7352	0.8574
No log	72.0	72	0.6382	0.6360	0.6380	0.7988
No log	73.0	73	0.5632	0.6534	0.5631	0.7504
No log	74.0	74	0.5539	0.6172	0.5538	0.7442
No log	75.0	75	0.5568	0.6166	0.5568	0.7462
No log	76.0	76	0.5665	0.6460	0.5665	0.7527
No log	77.0	77	0.6081	0.6506	0.6081	0.7798
No log	78.0	78	0.7184	0.6053	0.7182	0.8475
No log	79.0	79	0.8525	0.5644	0.8523	0.9232
No log	80.0	80	0.9258	0.5557	0.9255	0.9620
No log	81.0	81	0.9333	0.5566	0.9330	0.9659
No log	82.0	82	0.8978	0.5561	0.8975	0.9474
No log	83.0	83	0.8326	0.5612	0.8324	0.9123
No log	84.0	84	0.7476	0.5956	0.7474	0.8645
No log	85.0	85	0.6737	0.6491	0.6736	0.8207
No log	86.0	86	0.6470	0.6545	0.6469	0.8043
No log	87.0	87	0.6446	0.6545	0.6445	0.8028
No log	88.0	88	0.6600	0.6595	0.6599	0.8123
No log	89.0	89	0.6917	0.6320	0.6915	0.8316
No log	90.0	90	0.7170	0.6045	0.7168	0.8466
No log	91.0	91	0.7499	0.5940	0.7497	0.8658
No log	92.0	92	0.7887	0.5742	0.7885	0.8880
No log	93.0	93	0.8403	0.5616	0.8400	0.9165
No log	94.0	94	0.8664	0.5583	0.8661	0.9306
No log	95.0	95	0.8755	0.5570	0.8752	0.9355
No log	96.0	96	0.8700	0.5614	0.8697	0.9326
No log	97.0	97	0.8569	0.5587	0.8566	0.9255
No log	98.0	98	0.8410	0.5586	0.8407	0.9169
No log	99.0	99	0.8268	0.5645	0.8266	0.9092
No log	100.0	100	0.8190	0.5679	0.8187	0.9048

Framework versions

—Transformers 4.47.0
—Pytorch 2.5.1+cu121
—Datasets 3.2.0
—Tokenizers 0.21.0